TY - JOUR TI - Gradient boosting for hierarchical data in small area estimation AU - Messer, Paul AU - Schmid, Timo PY - 2026 SP - 1-29 PB - Otto-Friedrich-Universität CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/117558 LA - eng AB - Small Area Estimation (SAE) combines survey data with auxiliary sources such as administrative records, census data, or alternative datasets that typically offer broader coverage. By integrating these sources, SAE enhances the accuracy of (direct) survey estimates. To account for the hierarchical structure of survey data, model-based SAE methods often rely on Linear Mixed Models (LMMs). However, the distributional (e.g., normality) and structural (e.g., linearity) assumptions of LMMs may not always hold in practice, and the accuracy of model-based SAE depends on the validity of these assumptions. To address these limitations, we propose a Mixed Effect Gradient Boosting (MEGB) approach, which combines the flexibility of gradient boosting machines with the ability of mixed models to account for hierarchical data structures. MEGB extends standard gradient boosting by incorporating random effects, allowing it to capture unobserved heterogeneity across domains while retaining a nonparametric framework that models nonlinearities and interactions in the data. MEGB supports the derivation of area-level means from unit-level data and uses a nonparametric bootstrap to estimate the Mean Squared Error (MSE). Its performance is assessed through a model-based simulation study, comparing MEGB to established estimators, and further demonstrated using real-world data. The results suggest that MEGB offers promising area mean estimates and may outperform existing SAE methods in various scenarios. KW - Expectation–maximization algorithm KW - Ensemble learning KW - Mixed-effects modeling KW - Nonparametric bootstrap KW - Uncertainty quantification KW - Unit-level models ER - TY - JOUR TI - Gradient boosting for hierarchical data in small area estimation AU - Messer, Paul AU - Schmid, Timo PY - 2026 JO - Statistics and computing VL - 36 IS - 5, 218 SP - 1-29 PB - Springer Science + Business Media B.V CY - Dordrecht [u.a.] SN - 1573-1375 DO - 10.1007/s11222-026-10970-1 UR - https://fis.uni-bamberg.de/handle/uniba/117557 LA - eng AB - Small Area Estimation (SAE) combines survey data with auxiliary sources such as administrative records, census data, or alternative datasets that typically offer broader coverage. By integrating these sources, SAE enhances the accuracy of (direct) survey estimates. To account for the hierarchical structure of survey data, model-based SAE methods often rely on Linear Mixed Models (LMMs). However, the distributional (e.g., normality) and structural (e.g., linearity) assumptions of LMMs may not always hold in practice, and the accuracy of model-based SAE depends on the validity of these assumptions. To address these limitations, we propose a Mixed Effect Gradient Boosting (MEGB) approach, which combines the flexibility of gradient boosting machines with the ability of mixed models to account for hierarchical data structures. MEGB extends standard gradient boosting by incorporating random effects, allowing it to capture unobserved heterogeneity across domains while retaining a nonparametric framework that models nonlinearities and interactions in the data. MEGB supports the derivation of area-level means from unit-level data and uses a nonparametric bootstrap to estimate the Mean Squared Error (MSE). Its performance is assessed through a model-based simulation study, comparing MEGB to established estimators, and further demonstrated using real-world data. The results suggest that MEGB offers promising area mean estimates and may outperform existing SAE methods in various scenarios. KW - Expectation–maximization algorithm KW - Ensemble learning KW - Mixed-effects modeling KW - Nonparametric bootstrap KW - Uncertainty quantification KW - Unit-level models ER - TY - THES TI - Reversing the Coarse : Statistical Methods for Recovering Structure from Aggregated, Sparse, and Discretized Data AU - Mühlbauer, Michael PY - 2026 PB - Otto-Friedrich-Universität CY - Bamberg DO - 10.20378/irb-117096 UR - https://fis.uni-bamberg.de/handle/uniba/117096 LA - eng AB - Empirical research often works with data coarser than the question demands: a survey too sparse for district-level estimates, geocoordinates collapsed into regional totals, a continuous preference recorded on a five-point scale. This cumulative dissertation treats these as one problem — coarsened data — and recovers fine-grained structure by combining the coarse observation with auxiliary information. Chapter 1 estimates district-level person-kilometers traveled from the German household travel survey Mobilität in Deutschland 2017 with a log-transformed Fay-Herriot model across 1,590 domains. Covariates from official statistics and the INKAR database are selected in a three-phase LASSO procedure, reducing coefficients of variation by 16–44% against direct estimation. Chapter 2 extends kernel density estimation from aggregated data by integrating auxiliary spatial information through a correlation-based weighting scheme. The estimator is assessed in model-based simulations and in an evaluation study on population densities from the 2022 German Census in Bavaria, and then applied to brown hare hunting bag data in Lower Saxony. Chapter 3 applies the mixed-effects frequency-adjusted borders ordinal forest to lexical preference ratings in Maltese English and contributes partial dependence routines in R. It recovers curved, non-monotonic age trajectories that the cumulative link mixed model does not. KW - Coarsened data KW - Small area estimation KW - Fay-Herriot model KW - Kernel density estimation KW - Ordinal data KW - Random forests ER - TY - JOUR TI - Small area estimation with generalized random forests : estimating poverty rates in Mexico AU - Frink, Nicolas AU - Schmid, Timo AU - Würz, Nora PY - 2026 JO - Journal of the Royal Statistical Society: Series A, Statistics in society IS - Online-First PB - Oxford University Press CY - Oxford SN - 1467-985X DO - 10.1093/jrsssa/qnag092 UR - https://fis.uni-bamberg.de/handle/uniba/116849 LA - eng ER - TY - JOUR TI - Gender Roles and Career Paths : A Study of Ambitions and Work Preferences in Computer Science Undergraduates AU - Förtsch, Silvia Maria PY - 2026 JO - International Journal of Gender, Science and Technology VL - 18 IS - 1 SP - 171 - 199 PB - Open Univ. CY - Milton Keynes SN - 2040-0748 UR - https://fis.uni-bamberg.de/handle/uniba/116103 LA - eng ER - TY - JOUR TI - Calibrating Nonresponse Bias : A Cautionary Tale AU - Wildner, Raimund AU - Bosch, Volker AU - Meinfelder, Florian PY - 2026 SP - 292-311 PB - Otto-Friedrich-Universität CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/115542 LA - eng AB - Low response rates due to unit nonresponse have always been a ubiquitous problem in survey-based empirical research, and calibration is a popular method to adjust for bias caused by unit nonresponse. Typically, some external information on the true population quantities of margins for some calibration variables is available, and sometimes also of higher-order interactions. Weighting algorithms try to adjust the sample to these external benchmarks. It is generally assumed that even if the underlying missingness mechanism of the unit nonresponse is non-ignorable, weighting will at least alleviate the severity of the bias. We discuss data situations where weighting under a missing at random (MAR) assumption adjusts the sample correctly but still increases the bias for the analysis model, and we describe strategies for identifying auxiliary variables that are less susceptible to these unwanted effects. KW - unit nonresponse KW - calibration KW - weighting KW - selection at random KW - selection not at random ER - TY - JOUR TI - Calibrating Nonresponse Bias : A Cautionary Tale AU - Wildner, Raimund AU - Bosch, Volker AU - Meinfelder, Florian PY - 2026 JO - Journal of official statistics : JOS VL - 42 IS - 2 SP - 292-311 PB - SAGE CY - Stockholm SN - 2001-7367 DO - 10.1177/0282423x261425800 UR - https://fis.uni-bamberg.de/handle/uniba/115541 LA - eng AB - Low response rates due to unit nonresponse have always been a ubiquitous problem in survey-based empirical research, and calibration is a popular method to adjust for bias caused by unit nonresponse. Typically, some external information on the true population quantities of margins for some calibration variables is available, and sometimes also of higher-order interactions. Weighting algorithms try to adjust the sample to these external benchmarks. It is generally assumed that even if the underlying missingness mechanism of the unit nonresponse is non-ignorable, weighting will at least alleviate the severity of the bias. We discuss data situations where weighting under a missing at random (MAR) assumption adjusts the sample correctly but still increases the bias for the analysis model, and we describe strategies for identifying auxiliary variables that are less susceptible to these unwanted effects. KW - unit nonresponse KW - calibration KW - weighting KW - selection at random KW - selection not at random ER - TY - JOUR TI - Alternative selection mechanisms in online surveys AU - Prücklmair, Franz AU - Rendtel, Ulrich PY - 2026 SP - 165-193 PB - Otto-Friedrich-Universität CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/113134 LA - eng AB - The rapid changes in internet accessibility and the evolving ways it is used could raise doubts about whether earlier findings on selection effects in internet surveys still hold true. We address the following questions: 1) Is the selection criterion by mere internet access still a reasonable self-selection criterion? 2) If this is no longer true, how should alternative self-selection processes be modeled? 3) How can these selection processes be controlled? Are demographic control variables sufficient to establish the Missing at Random (MAR) condition, or is it possible to establish the MAR condition with more powerful control variables? 4) To what extent do the weighting procedures correct a self-selection bias? We investigate these questions in the setting of a simulation study where we assume four different selection models. These involve the length of internet use, posting behavior, and interest in politics and are based on theoretical considerations. We use the European Social Survey (ESS) as a simulation environment, which contains these variables and demographic background variables. It also includes our outcome variable, the vote in the 2017 Bundestag election in Germany. In order to judge the differences of the non-probability results and the simulated universe we compare the results of ESS estimates of the 2017 Bundestags elections the real election results. KW - Online surveys KW - River Sampling KW - Non-probability sampl KW - Quasi-Randomization approach KW - Calibration KW - European Social Survey KW - Online-Umfragen KW - Nicht-Zufallsstichproben KW - Quasi-Randomisierung KW - Kalibrierung ER - TY - JOUR TI - Analysing Opportunity Cost of Care Work using Mixed Effects Random Forests under Aggregated Auxiliary Data AU - Krennmair, Patrick AU - Würz, Nora AU - Schmid, Timo PY - 2026 JO - Journal of the Royal Statistical Society. Series C, Applied statistics VL - 75 IS - 1 SP - 1-20 PB - Oxford University Press CY - Kettering ; Cary, NC ; Tokyo SN - 1467-9876 DO - 10.1093/jrsssc/qlaf031 UR - https://fis.uni-bamberg.de/handle/uniba/108854 LA - eng ER - TY - JOUR TI - Alternative selection mechanisms in online surveys AU - Prücklmair, Franz AU - Rendtel, Ulrich PY - 2025 JO - Wirtschafts- und sozialstatistisches Archiv : ASTA ; eine Zeitschrift der Deutschen Statistischen Gesellschaft VL - 19 IS - 3-4 SP - 165-193 PB - Springer CY - Berlin ; Heidelberg SN - 1863-8163 DO - 10.1007/s11943-025-00362-8 UR - https://fis.uni-bamberg.de/handle/uniba/113103 LA - eng AB - The rapid changes in internet accessibility and the evolving ways it is used could raise doubts about whether earlier findings on selection effects in internet surveys still hold true. We address the following questions: 1) Is the selection criterion by mere internet access still a reasonable self-selection criterion? 2) If this is no longer true, how should alternative self-selection processes be modeled? 3) How can these selection processes be controlled? Are demographic control variables sufficient to establish the Missing at Random (MAR) condition, or is it possible to establish the MAR condition with more powerful control variables? 4) To what extent do the weighting procedures correct a self-selection bias? We investigate these questions in the setting of a simulation study where we assume four different selection models. These involve the length of internet use, posting behavior, and interest in politics and are based on theoretical considerations. We use the European Social Survey (ESS) as a simulation environment, which contains these variables and demographic background variables. It also includes our outcome variable, the vote in the 2017 Bundestag election in Germany. In order to judge the differences of the non-probability results and the simulated universe we compare the results of ESS estimates of the 2017 Bundestags elections the real election results. KW - Online surveys KW - River Sampling KW - Non-probability sampl KW - Quasi-Randomization approach KW - Calibration KW - European Social Survey KW - Online-Umfragen KW - Nicht-Zufallsstichproben KW - Quasi-Randomisierung KW - Kalibrierung ER - TY - JOUR TI - Small Area Estimation of Poverty in Four West African Countries by Integrating Survey and Geospatial Data AU - Edochie, Ifeanyi AU - Newhouse, David AU - Tzavidis, Nikos AU - Schmid, Timo AU - Foster, Elizabeth AU - Hernandez, Angela Luna AU - Ouedraogo, Aissatou AU - Sanoh, Aly AU - Savadogo, Aboudrahyme PY - 2025 SP - 96-124 PB - Otto-Friedrich-Universität CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/111678 LA - eng AB - The paper presents methodology to generate experimental small area estimates (SAE) of poverty in four West African countries: Chad, Guinea, Mali, and Niger. Due to the absence of recent census data in the four countries, household level survey data are integrated with grid-level geospatial data, which are used as covariates in model-based estimation. Leveraging geospatial data enables reporting of poverty estimates more frequently at disaggregated administrative levels and makes estimation feasible in areas for which survey data are not available. The paper leverages the availability of a recent census in Burkina Faso for evaluation purposes. Estimates obtained with the same survey instruments and candidate geospatial covariates as the other four countries are compared against estimates obtained using recent census data and an empirical best predictor under a unit level model. For Burkina Faso, estimates obtained using geospatial data are highly correlated with the census-based ones in sampled areas but moderately correlated in non-sampled areas. The results demonstrate that in the absence of recent census data, small area estimation with publicly available geospatial covariates is feasible, can lead to large efficiency improvements compared to direct estimation, and improve the timeliness of small area estimates. KW - headcount poverty rates KW - nested error regression model KW - geospatial covariates KW - empirical best predictor ER - TY - JOUR TI - Validating an Index of Selection Bias for Proportions in Non‐Probability Samples AU - Hammon, Angelina AU - Zinn, Sabine PY - 2025 SP - 499-516 PB - Otto-Friedrich-Universität CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/111357 LA - eng AB - Fast online surveys without sampling frames are becoming increasingly important in survey research. Their recruitment methods result in non-probability samples. As the mechanism of data generation is always unknown in such samples, the problem of non-ignorability arises making vgeneralisation of calculated statistics to the population of interest highly questionable. Sensitivity analyses provide a valuable tool to deal with non-ignorability. They capture the impact of different sample selection mechanisms on target statistics. In 2019, Andridge and colleagues proposed an index to quantify potential (non-ignorable) selection bias in proportions that combines the effects of different selection mechanisms. In this paper, we validate this index with an artificial non-probability sample generated from a large empirical data set and additionally applied it to proportions estimated from data on current political attitudes arising from a real non-probability sample selected via River sampling. We find a number of conditions that must be met for the index to perform meaningfully. When these requirements are fulfilled, the index shows an overall good performance in both of our applications in detecting and correcting present selection bias in estimated proportions. Thus, it provides a powerful measure for evaluating the robustness of results obtained from non-probability samples. KW - non-ignorable sample selection KW - pattern-mixture modelling KW - selection not at random KW - sensitivity analysis ER - TY - JOUR TI - Experimental UK Regional Consumer Price Inflation with Model-Based Expenditure Weights AU - Dawber, James AU - Würz, Nora AU - Smith, Paul A. AU - Flower, Tanya AU - Thomas, Heledd AU - Schmid, Timo AU - Tzavidis, Nikos PY - 2025 SP - 213–237 PB - Otto-Friedrich-Universität CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/110536 LA - deu AB - Like many other countries, the United Kingdom (UK) produces a national consumer price index (CPI) to measure inflation. Presently, CPI measures are not produced for regions within the UK. It is believed that, using only available data sources, a regional CPI would not be precise or reliable enough as an official statistic, primarily because the regional partitioning of the data makes sample sizes too small. We investigate this claim by producing experimental regional CPIs using publicly available price data, and deriving expenditure weights from the Living Costs and Food survey. We detail the methods and challenges of developing a regional CPI and evaluate its reliability. We then assess whether model-based methods such as smoothing and small area estimation significantly improve the measures. We find that a regional CPI can be produced with available data sources, however it appears to be excessively volatile over time, mainly due to the weights. Smoothing and small area estimation improve the reliability of the regional CPI series to some extent but they remain too volatile for regional policy use. This research provides a valuable framework for the development of a more viable regional CPI measure for the UK in the future. KW - CPI conceptual framework KW - basket of goods and services KW - small area estimation KW - Fay-Herriot models ER - TY - JOUR TI - Small Area with Multiply Imputed Survey Data AU - Runge, Marina AU - Schmid, Timo PY - 2025 SP - 507-533 PB - Otto-Friedrich-Universität CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/110603 LA - eng AB - In this article, we propose a framework for small area estimation with multiply imputed survey data. Many statistical surveys suffer from (a) high nonresponse rates due to sensitive questions and response burden and (b) too small sample sizes to allow for reliable estimates on (unplanned) disaggregated levels due to budget constraints. One way to deal with missing values is to replace them by several plausible/imputed values based on a model. Small area estimation, such as the model by Fay and Herriot, is applied to estimate regionally disaggregated indicators when direct estimates are imprecise. The framework presented tackles simultaneously multiply imputed values and imprecise direct estimates. In particular, we extend the general class of transformed Fay-Herriot models to account for the additional uncertainty from multiple imputation. We derive three special cases of the Fay-Herriot model with particular transformations and provide point and mean squared error estimators. Depending on the case, the mean squared error is estimated by analytic solutions or resampling methods. Comprehensive simulations in a controlled environment show that the proposed methodology leads to reliable and precise results in terms of bias and mean squared error. The methodology is illustrated by a real data example using European wealth data. KW - Fay-Herriot model KW - mean squared error KW - multiple imputation KW - nonresponse KW - survey statistics ER - TY - JOUR TI - Editorial AU - Schmid, Timo AU - Zwick, Markus PY - 2025 SP - 59–61 PB - Otto-Friedrich-Universität CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/110314 LA - deu AB - - KW - - ER - TY - JOUR TI - Editorial AU - Zwick, Markus AU - Schmid, Timo PY - 2025 SP - 1-3 PB - Otto-Friedrich-Universität CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/110313 LA - deu AB - - KW - - ER - TY - JOUR TI - Small area prediction of counts under machine learning-type mixed models AU - Frink, Nicolas AU - Schmid, Timo PY - 2025 SP - 1-16 PB - Otto-Friedrich-Universität CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/109919 LA - eng AB - Small area estimation methods are proposed that use generalized tree-based machine learning techniques to improve the estimation of disaggregated means in small areas using discrete survey data. Specifically, two existing approaches based on random forests - the Generalized Mixed Effects Random Forest (GMERF) and a Mixed Effects Random Forest (MERF) - are extended to accommodate count outcomes, addressing key challenges such as overdispersion. Additionally, three bootstrap methodologies designed to assess the reliability of point estimators for area-level means are evaluated. The numerical analysis shows that the MERF, which does not assume a Poisson distribution to model the mean behavior of count data, excels in scenarios of severe overdispersion. Conversely, the GMERF performs best under conditions where Poisson distribution assumptions are moderately met. In a case study using real-world data from the state of Guerrero, Mexico, the proposed methods effectively estimate area-level means while capturing the uncertainty inherent in overdispersed count data. These findings highlight their practical applicability for small area estimation. KW - Bootstrap KW - Generalized linear mixed models KW - Overdispersed count data KW - Random forest KW - Small area estimation ER - TY - JOUR TI - Releasing survey microdata with exact cluster locations and additional privacy safeguards AU - Koebe, Till AU - Arias‐Salazar, Alejandra AU - Schmid, Timo PY - 2025 SP - 1-13 PB - Otto-Friedrich-Universität CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/109866 LA - eng AB - Household survey programs around the world publish fine-granular georeferenced microdata to support research on the interdependence of human livelihoods and their surrounding environment. To safeguard the respondents’ privacy, micro-level survey data is usually (pseudo)-anonymized through deletion or perturbation procedures such as obfuscating the true location of data collection. This, however, poses a challenge to emerging approaches that augment survey data with auxiliary information on a local level. Here, we propose an alternative microdata dissemination strategy that leverages the utility of the original microdata with additional privacy safeguards through synthetically generated data using generative models. We back our proposal with experiments using data from the 2011 Costa Rican census and satellite-derived auxiliary information. Our strategy reduces the respondents’ re-identification risk for any number of disclosed attributes by 60–80% even under re-identification attempts. KW - survey microdata ER - TY - CPAPER TI - Yes you can, follow your goals! : Individual Coaching for female Computer scientists on career development AU - Förtsch, Silvia PY - 2025 SP - 2603-5871 PB - Otto-Friedrich-Universität CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/109608 LA - eng AB - Earlier results show that female computer scientists show a lower self-efficacy compared to male colleagues. Similarly, gender-typical attributions, in the sense of gender stereotypes, are considered a disorder. As a result, a coaching program that supports computer scientists after a re-entry into professional life also, in a new orientation or with regard to management ambitions has been developed at the University of . The measure strengthens computer scientists in their motivational resources and enables them to take advantage of professional opportunities. A scientifically founded analysis of their potential helps the coachees to become aware of their abilities and competences. The coaching program based on the potential analysis, takes important life goals of the coachees into account. Individual career plans are developed in coaching sessions, including a clear definition of the objective and implementation strategy. An open and modern corporate culture offers new opportunities of career paths for female computer scientists. If the company philosophy is based on appreciation of good performance, women in computer science are more motivated strengthen their career ambitions. KW - women in computer science; KW - self-efficacy beliefs KW - potential analysis KW - individual coaching KW - career development ER - TY - JOUR TI - Small area prediction of counts under machine learning-type mixed models AU - Frink, Nicolas AU - Schmid, Timo PY - 2025 JO - Computational Statistics & Data Analysis VL - 211 IS - 108218 SP - 1-16 PB - Elsevier Science CY - Amsterdam SN - 0167-9473 DO - 10.1016/j.csda.2025.108218 UR - https://fis.uni-bamberg.de/handle/uniba/108853 LA - eng AB - Small area estimation methods are proposed that use generalized tree-based machine learning techniques to improve the estimation of disaggregated means in small areas using discrete survey data. Specifically, two existing approaches based on random forests - the Generalized Mixed Effects Random Forest (GMERF) and a Mixed Effects Random Forest (MERF) - are extended to accommodate count outcomes, addressing key challenges such as overdispersion. Additionally, three bootstrap methodologies designed to assess the reliability of point estimators for area-level means are evaluated. The numerical analysis shows that the MERF, which does not assume a Poisson distribution to model the mean behavior of count data, excels in scenarios of severe overdispersion. Conversely, the GMERF performs best under conditions where Poisson distribution assumptions are moderately met. In a case study using real-world data from the state of Guerrero, Mexico, the proposed methods effectively estimate area-level means while capturing the uncertainty inherent in overdispersed count data. These findings highlight their practical applicability for small area estimation. KW - Bootstrap KW - Generalized linear mixed models KW - Overdispersed count data KW - Random forest KW - Small area estimation ER - TY - JOUR TI - Kernel density smoothing of composite spatial data on administrative area level AU - Erfurth, Kerstin AU - Groß, Marcus AU - Rendtel, Ulrich AU - Schmid, Timo PY - 2025 SP - 25-49 PB - Otto-Friedrich-Universität CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/106603 LA - eng AB - Composite spatial data on administrative area level are often presented by maps. The aim is to detect regional differences in the concentration of subpopulations, like elderly persons, ethnic minorities, low-educated persons, voters of a political party or persons with a certain disease. Thematic collections of such maps are presented in different atlases. The standard presentation is by Choropleth maps where each administrative unit is represented by a single value. These maps can be criticized under three aspects: the implicit assumption of a uniform distribution within the area, the instability of the resulting map with respect to a change of the reference area and the discontinuities of the maps at the borderlines of the reference areas which inhibit the detection of regional clusters. In order to address these problems we use a density approach in the construction of maps. This approach does not enforce a local uniform distribution. It does not depend on a specifc choice of area reference system and there are no discontinuities in the displayed maps. A standard estimation procedure of densities are Kernel density estimates. However, these estimates need the geo-coordinates of the single units which are not at disposal as we have only access to the aggregates of some area system. To overcome this hurdle, we use a statistical simulation concept. This can be interpreted as a Simulated Expectation Maximisation (SEM) algorithm of Celeux et al (1996). We simulate observations from the current density estimates which are consistent with the aggregation information (S-step). Then we apply the Kernel density estimator to the simulated sample which gives the next density estimate (E-Step). This concept has been f rst applied for grid data with rectangular areas, see Groß et al (2017), for the display of ethnic minorities. In a second application we demonstrated the use of this approach for the so-called “change of support” (Bradley et al 2016) problem. Here Groß et al (2020) used the SEM algorithm to recalculate case numbers between non-hierarchical administrative area systems. Recently Rendtel et al (2021) applied the SEM algorithm to display spatial-temporal clusters of Corona infections in Germany. Here we present three modif cations of the basic SEM algorithm: 1) We introduce a boundary correction which removes the underestimation of kernel density estimates at the borders of the population area. 2) We recognize unsettled areas, like lakes, parks and industrial areas, in the computation of the kernel density. 3) We adapt the SEM algorithm for the computation of local percentages which are important especially in voting analysis. We evaluate our approach against several standard maps by means of the local voting register with known addresses. In the empirical part we apply our approach for the display of voting results for the 2016 election of the Berlin parliament. We contrast our results against Choropleth maps and show new possibilities for reporting spatial voting results. KW - Spatial data KW - Administrative areas KW - Choropleths KW - Kernel density estimation KW - Voting atlases ER - TY - JOUR TI - Intercensal updating using structure‐preserving methods and satellite imagery AU - Koebe, Till AU - Arias‐Salazar, Alejandra AU - Rojas‐Perilla, Natalia AU - Schmid, Timo PY - 2025 SP - 1-27 PB - Otto-Friedrich-Universität CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/106925 LA - eng AB - Censuses are fundamental building blocks of most modern-day societies, yet collected every 10 years at best. We propose an extension of the widely popular census updating technique structure-preserving estimation by incorporating auxiliary information in order to take ongoing subnational population shifts into account. We apply our method by incorporating satellite imagery as additional source to derive annual small-area updates of multidimensional poverty indicators from 2013 to 2020 for a population at risk: female-headed households in Senegal. We evaluate the performance of our proposal using data from two different census periods. KW - multidimensional poverty KW - official statistics KW - small-area estimation KW - SPREE ER - TY - JOUR TI - Validating an Index of Selection Bias for Proportions in Non‐Probability Samples AU - Hammon, Angelina AU - Zinn, Sabine PY - 2025 JO - International statistical review : a journal of the International Statistical Institute and its associations VL - 93 IS - 3 SP - 499-516 PB - Wiley-Blackwell CY - Oxford ; Malden, Mass. SN - 0020-8779 DO - 10.1111/insr.12590 UR - https://fis.uni-bamberg.de/handle/uniba/107715 LA - eng AB - Fast online surveys without sampling frames are becoming increasingly important in survey research. Their recruitment methods result in non-probability samples. As the mechanism of data generation is always unknown in such samples, the problem of non-ignorability arises making vgeneralisation of calculated statistics to the population of interest highly questionable. Sensitivity analyses provide a valuable tool to deal with non-ignorability. They capture the impact of different sample selection mechanisms on target statistics. In 2019, Andridge and colleagues proposed an index to quantify potential (non-ignorable) selection bias in proportions that combines the effects of different selection mechanisms. In this paper, we validate this index with an artificial non-probability sample generated from a large empirical data set and additionally applied it to proportions estimated from data on current political attitudes arising from a real non-probability sample selected via River sampling. We find a number of conditions that must be met for the index to perform meaningfully. When these requirements are fulfilled, the index shows an overall good performance in both of our applications in detecting and correcting present selection bias in estimated proportions. Thus, it provides a powerful measure for evaluating the robustness of results obtained from non-probability samples. KW - non-ignorable sample selection KW - pattern-mixture modelling KW - selection not at random KW - sensitivity analysis ER - TY - JOUR TI - Small Area Estimation of Poverty in Four West African Countries by Integrating Survey and Geospatial Data AU - Edochie, Ifeanyi AU - Newhouse, David AU - Tzavidis, Nikos AU - Schmid, Timo AU - Foster, Elizabeth AU - Hernandez, Angela Luna AU - Ouedraogo, Aissatou AU - Sanoh, Aly AU - Savadogo, Aboudrahyme PY - 2025 JO - Journal of Official Statistics VL - 41 IS - 1 SP - 96-124 PB - Sage CY - London SN - 2001-7367 DO - 10.1177/0282423x241284890 UR - https://fis.uni-bamberg.de/handle/uniba/105797 LA - eng AB - The paper presents methodology to generate experimental small area estimates (SAE) of poverty in four West African countries: Chad, Guinea, Mali, and Niger. Due to the absence of recent census data in the four countries, household level survey data are integrated with grid-level geospatial data, which are used as covariates in model-based estimation. Leveraging geospatial data enables reporting of poverty estimates more frequently at disaggregated administrative levels and makes estimation feasible in areas for which survey data are not available. The paper leverages the availability of a recent census in Burkina Faso for evaluation purposes. Estimates obtained with the same survey instruments and candidate geospatial covariates as the other four countries are compared against estimates obtained using recent census data and an empirical best predictor under a unit level model. For Burkina Faso, estimates obtained using geospatial data are highly correlated with the census-based ones in sampled areas but moderately correlated in non-sampled areas. The results demonstrate that in the absence of recent census data, small area estimation with publicly available geospatial covariates is feasible, can lead to large efficiency improvements compared to direct estimation, and improve the timeliness of small area estimates. KW - headcount poverty rates KW - nested error regression model KW - geospatial covariates KW - empirical best predictor ER - TY - GEN TI - On the Improvement of Predictive Modeling Using Bayesian Stacking and Posterior Predictive Checking AU - Nold, Mariana Saskia AU - Meinfelder, Florian AU - Kaplan, David PY - 2024 SP - 1-40 PB - arXiv DO - 10.48550/arxiv.2402.19046 UR - https://fis.uni-bamberg.de/handle/uniba/116858 LA - eng ER - TY - JOUR TI - A Modern Approach to Stability Studies via Bayesian Linear Mixed Models Incorporating Auxiliary Effects AU - Cordero Alba, Miguel Angel AU - Meinfelder, Florian AU - Eilert, Tobias PY - 2024 JO - Journal of Pharmaceutical Sciences VL - 113 IS - 7 SP - 1779-1793 PB - Elsevier CY - Amsterdam SN - 1520-6017 DO - 10.1016/j.xphs.2024.02.020 UR - https://fis.uni-bamberg.de/handle/uniba/104632 LA - eng ER - TY - THES TI - Recent Advances in Small Area Estimation of Economic and Poverty Indicators using Traditional and Alternative Data AU - Lee, Yeonjoo PY - 2024 PB - Otto-Friedrich-Universität CY - Bamberg DO - 10.20378/irb-103063 UR - https://fis.uni-bamberg.de/handle/uniba/103063 LA - eng AB - Chapter 1 - Variable selection using conditional AIC for linear mixed models with data-driven transformations When data analysts use linear mixed models, they usually encounter two practical problems: a) the true model is unknown and b) the Gaussian assumptions of the errors do not hold. While these problems commonly appear together, researchers tend to treat them individually by a) finding an optimal model based on the conditional Akaike information criterion (cAIC) and b) applying transformations on the dependent variable. However, the optimal model depends on the transformation and vice versa. In this paper, we aim to solve both problems simultaneously. In particular, we propose an adjusted cAIC by using the Jacobian of the particular transformation such that various model candidates with differently transformed data can be compared. From a computational perspective, we propose a step-wise selection approach based on the introduced adjusted cAIC. Model-based simulations are used to compare the proposed selection approach to alternative approaches. Finally, the introduced approach is applied to Mexican data to estimate poverty and inequality indicators for 81 municipalities. KW - Box-Cox transformation KW - Empirical best predictor KW - Indicators KW - Small area estimation KW - inflation KW - regional inequality KW - regional difference KW - price statistics KW - household consumption KW - head count ratio KW - Mozambique KW - satellite imagery KW - remote sensing data KW - subarea level model ER - TY - JOUR TI - Estimating regional unemployment with mobile network data for Functional Urban Areas in Germany AU - Hadam, Sandra AU - Würz, Nora AU - Kreutzmann, Ann-Kristin AU - Schmid, Timo PY - 2024 SP - 205-233 PB - Otto-Friedrich-Universität CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/98448 LA - eng AB - The ongoing growth of cities due to better job opportunities is leading to increased labour-related commuter flows in several countries. On the one hand, an increasing number of people commute and move to the cities, but on the other hand, the labour market indicates higher unemployment rates in urban areas than in the surrounding areas. We investigate this phenomenon on regional level by an alternative definition of unemployment rates in which commuting behaviour is integrated. We combine data from the Labour Force Survey with dynamic mobile network data by small area models for the federal state North Rhine-Westphalia in Germany. From a methodical perspective, we use a transformed Fay–Herriot model with bias correction for the estimation of unemployment rates and propose a parametric bootstrap for the mean squared error estimation that includes the bias correction. The performance of the proposed methodology is evaluated in a case study based on official data and in model-based simulations. The results in the application show that unemployment rates (adjusted by commuters) in German cities are lower than traditional official unemployment rates indicate. KW - Bias correction KW - Fay–Herriot model KW - Mean squared error KW - Small area estimation KW - Unemployment rates ER - TY - JOUR TI - Eine Analyse des Studienerfolgs im Masterstudium auf der Basis von Umfrage- und administrativen Prüfungsdaten : Ein Vergleich von fünf Masterstudiengängen am Fachbereich Wirtschaftswissenschaft der Freien Universität Berlin AU - Rendtel, Ulrich AU - Lee, Yeonjoo AU - Gerks, Hartmut PY - 2024 SP - 37-76 PB - Otto-Friedrich-Universität CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/95689 LA - deu AB - In einer Umfrage unter den neu immatrikulierten Master-Studierenden wurden unter anderem der soziale Hintergrund, die Studienfinanzierung und die Motivation für den gewählten Studiengang erfragt. Durch die Verknüpfung von administrativen Prüfungsdaten mit den neuen Umfragedaten wird die Analyse der Studienverläufe und des Studienerfolgs im Zusammenhang mit individuellen Merkmalen ermöglicht. Damit vermeiden wir die Schwachstellen üblicher studentischer Befragungen mit hohem Non-response und Erinnerungsfehlern. Dieser Ansatz wird hier zum Vergleich von fünf Masterstudiengängen am Fachbereich (FB) Wirtschaftswissenschaft der Freien Universität Berlin (FU Berlin) benutzt. Nach der ersten Befragung wurden die Studierenden über die ersten 6 Fachsemester hinsichtlich des Erwerbs von Leistungspunkten (LP), dem Abschluss ihres Masterstudiums sowie den dabei erzielten Noten begleitet. In den verknüpften Daten konnte ein Erfolgsbias der teilnahmebereiten Studierenden der Kohorte festgestellt werden, der aber durch eine geeignete Gewichtung über die Responserate gut beherrschbar ist. Wir vergleichen die Studiengänge in verschiedenen Phasen: Studieneingangsphase, Erreichen der Regelstudienzeit und Abschluss bis zum 6 Fachsemester. Weiterhin wird der Einfluss von Hintergrundmerkmalen auf die erreichte Note beim Studienabschluss untersucht. In der Studieneingangsphase ergeben sich deutliche Unterschiede zwischen den Studiengängen. Konditioniert man allerdings auf den Studienerfolg im ersten Semester, so verschwinden diese Unterschiede für die zweite Studienphase und den Abschluss des Studiums. Überraschend ist der geringe Einfluss der Bachelor-Note auf den Studienerfolg und die erzielte Master-Note. Die Ergebnisse zeigen die Möglichkeit auf, einen möglichen Studienabbruch schon relativ sicher anhand der erzielten LP in der Studieneingangsphase zu erkennen. KW - Administrative Prüfungsdaten KW - Umfragedaten KW - Studienerfolg KW - Master Programme KW - Studiengang Monitoring KW - Administrative Examination Data KW - Survey Data KW - Study Success KW - Master Programs KW - Study Program Monitoring ER - TY - JOUR TI - A Framework for Producing Small Area Estimates Based on Area-Level Models in R AU - Harmening, Sylvia AU - Kreutzmann, Ann-Kristin AU - Schmidt, Sören AU - Salvati, Nicola AU - Schmid, Timo PY - 2024 SP - 316-341 PB - Otto-Friedrich-Universität CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/94974 LA - eng AB - The R package emdi facilitates the estimation of regionally disaggregated indicators using small area estimation methods and provides tools for model building, diagnostics, presenting, and exporting the results. The package version 1.1.7 includes unit-level small area models that rely on access to micro data. The area-level model by Fay and Herriot (1979) and various extensions have been added to the package since the release of version 2.0.0. These extensions include (a) area-level models with back-transformations, (b) spatial and robust extensions, (c) adjusted variance estimation methods, and (d) area-level models that account for measurement errors. Corresponding mean squared error estimators are implemented for assessing the uncertainty. User-friendly tools like a stepwise variable selection, model diagnostics, benchmarking options, high quality maps and results exportation options enable a complete analysis procedure. The functionality of the package is illustrated by examples based on synthetic data for Austrian districts. KW - Area-Level Models in R ER - TY - GEN TI - Latent-variable modeling of ordinal outcomes in language data analysis AU - Sönning, Lukas AU - Krug, Manfred AU - Vetter, Fabian AU - Schmid, Timo AU - Leucht, Anne AU - Messer, Paul PY - 2024 SP - 1-22 PB - Otto-Friedrich-Universität CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/95382 LA - eng AB - In empirical work, ordinal variables are typically analyzed using means based on numeric scores assigned to categories. While this strategy has met with justified criticism in the methodological literature, it also generates simple and informative data summaries, a standard often not met by statistically more adequate procedures. Motivated by a survey of how ordered variables are dealt with in language research, we draw attention to an un(der)used latent-variable approach to ordinal data modeling, an alternative perspective on the most widely used form of ordered regression, the cumulative model. Since the latent-variable approach is not mentioned in statistical textbooks for linguists and does not feature in any of the studies in our survey, we believe it is worthwhile to promote its benefits. To this end, we draw on questionnaire-based preference ratings by speakers of Maltese English, who indicated on a 5-point scale which of two synonymous expressions (e.g. package-parcel) they (tend to) use. We demonstrate that a latent-variable analysis affords nuanced and interpretable data summaries that can be visualized effectively, while at the same time avoiding limitations inherent in mean response models (e.g. distortions induced by floor and ceiling effects). The online supplementary materials include a tutorial for its implementation in R. KW - Latent-variable modeling ER - TY - JOUR TI - Latent-Variable Modelling of Ordinal Outcomes in Language Data Analysis AU - Sönning, Lukas AU - Krug, Manfred AU - Vetter, Fabian AU - Schmid, Timo AU - Leucht, Anne AU - Messer, Paul PY - 2024 JO - Journal of quantitative linguistics : official journal of the International Quantitative Linguistics Association (IQLA) VL - 31 IS - 2 SP - 77-106 PB - Informa UK Limited CY - London SN - 1744-5035 DO - 10.1080/09296174.2024.2329448 UR - https://fis.uni-bamberg.de/handle/uniba/95412 LA - eng ER - TY - JOUR TI - A Bayesian Approach Towards Missing Covariate Data in Multilevel Latent Regression Models AU - Aßmann, Christian AU - Gaasch, Jean-Christoph AU - Stingl, Doris PY - 2024 SP - 1495-1528 PB - Otto-Friedrich-Universität CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/93601 LA - eng AB - The measurement of latent traits and investigation of relations between these and a potentially large set of explaining variables is typical in psychology, economics, and the social sciences. Corresponding analysis often relies on surveyed data from large-scale studies involving hierarchical structures and missing values in the set of considered covariates. This paper proposes a Bayesian estimation approach based on the device of data augmentation that addresses the handling of missing values in multilevel latent regression models. Population heterogeneity is modeled via multiple groups enriched with random intercepts. Bayesian estimation is implemented in terms of a Markov chain Monte Carlo sampling approach. To handle missing values, the sampling scheme is augmented to incorporate sampling from the full conditional distributions of missing values. We suggest to model the full conditional distributions of missing values in terms of non-parametric classification and regression trees. This offers the possibility to consider information from latent quantities functioning as sufficient statistics. A simulation study reveals that this Bayesian approach provides valid inference and outperforms complete cases analysis and multiple imputation in terms of statistical efficiency and computation time involved. An empirical illustration using data on mathematical competencies demonstrates the usefulness of the suggested approach. KW - Item response theory KW - population heterogeneity KW - Markov chain Monte Carlo KW - classification and regression trees KW - missing values ER - TY - JOUR TI - Multiple imputation of ordinal missing not at random data AU - Hammon, Angelina PY - 2024 SP - 671-692 PB - Otto-Friedrich-Universität CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/93529 LA - eng AB - We introduce a selection model-based imputation approach to be used within the Fully Conditional Specification (FCS) framework for the Multiple Imputation (MI) of incomplete ordinal variables that are supposed to be Missing Not at Random (MNAR). Thereby, we generalise previous work on this topic which involved binary single-level and multilevel data to ordinal variables. We apply an ordered probit model with sample selection as base of our imputation algorithm. The applied model involves two equations that are modelled jointly where the first one describes the missing-data mechanism and the second one specifies the variable to be imputed. In addition, we develop a version for hierarchical data by incorporating random intercept terms in both equations. To fit this multilevel imputation model we use quadrature techniques. Two simulation studies validate the overall good performance of our single-level and multilevel imputation methods. In addition, we show its applicability to empirical data by applying it to a common research topic in educational science using data of the National Educational Panel Study (NEPS) and conducting a short sensitivity analysis. Our approach is designed to be used within the R software package mice which makes it easy to access and apply. KW - Fully conditional specification KW - Missingness not at random KW - Multilevel data KW - Multiple imputation KW - Selection model KW - Ordinal data ER - TY - GEN TI - Latent-variable modeling of ordinal outcomes in language data analysis AU - Sönning, Lukas AU - Krug, Manfred AU - Vetter, Fabian AU - Schmid, Timo AU - Leucht, Anne AU - Messer, Paul PY - 2024 SP - 1-22 PB - Center for Open Science DO - 10.31219/osf.io/jhv6b UR - https://fis.uni-bamberg.de/handle/uniba/93359 LA - eng AB - In empirical work, ordinal variables are typically analyzed using means based on numeric scores assigned to categories. While this strategy has met with justified criticism in the methodological literature, it also generates simple and informative data summaries, a standard often not met by statistically more adequate procedures. Motivated by a survey of how ordered variables are dealt with in language research, we draw attention to an un(der)used latent-variable approach to ordinal data modeling, an alternative perspective on the most widely used form of ordered regression, the cumulative model. Since the latent-variable approach is not mentioned in statistical textbooks for linguists and does not feature in any of the studies in our survey, we believe it is worthwhile to promote its benefits. To this end, we draw on questionnaire-based preference ratings by speakers of Maltese English, who indicated on a 5-point scale which of two synonymous expressions (e.g. package-parcel) they (tend to) use. We demonstrate that a latent-variable analysis affords nuanced and interpretable data summaries that can be visualized effectively, while at the same time avoiding limitations inherent in mean response models (e.g. distortions induced by floor and ceiling effects). The online supplementary materials include a tutorial for its implementation in R. KW - Latent-variable modeling ER - TY - JOUR TI - Estimating regional unemployment with mobile network data for Functional Urban Areas in Germany AU - Hadam, Sandra AU - Würz, Nora AU - Kreutzmann, Ann-Kristin AU - Schmid, Timo PY - 2024 JO - Statistical methods & applications VL - 33 IS - 1 SP - 205-233 PB - Springer CY - Berlin SN - 1618-2510 DO - 10.1007/s10260-023-00722-0 UR - https://fis.uni-bamberg.de/handle/uniba/92359 LA - eng AB - The ongoing growth of cities due to better job opportunities is leading to increased labour-related commuter flows in several countries. On the one hand, an increasing number of people commute and move to the cities, but on the other hand, the labour market indicates higher unemployment rates in urban areas than in the surrounding areas. We investigate this phenomenon on regional level by an alternative definition of unemployment rates in which commuting behaviour is integrated. We combine data from the Labour Force Survey with dynamic mobile network data by small area models for the federal state North Rhine-Westphalia in Germany. From a methodical perspective, we use a transformed Fay–Herriot model with bias correction for the estimation of unemployment rates and propose a parametric bootstrap for the mean squared error estimation that includes the bias correction. The performance of the proposed methodology is evaluated in a case study based on official data and in model-based simulations. The results in the application show that unemployment rates (adjusted by commuters) in German cities are lower than traditional official unemployment rates indicate. KW - Bias correction KW - Fay–Herriot model KW - Mean squared error KW - Small area estimation KW - Unemployment rates ER - TY - JOUR TI - Eine Analyse des Studienerfolgs im Masterstudium auf der Basis von Umfrage- und administrativen Prüfungsdaten : Ein Vergleich von fünf Masterstudiengängen am Fachbereich Wirtschaftswissenschaft der Freien Universität Berlin AU - Rendtel, Ulrich AU - Lee, Yeonjoo AU - Gerks, Hartmut PY - 2024 JO - AStA Wirtschafts- und Sozialstatistisches Archiv VL - 18 IS - 1 SP - 37-76 PB - Springer CY - Berlin ; Heidelberg SN - 1863-8155 DO - 10.1007/s11943-023-00325-x UR - https://fis.uni-bamberg.de/handle/uniba/90267 LA - deu AB - In einer Umfrage unter den neu immatrikulierten Master-Studierenden wurden unter anderem der soziale Hintergrund, die Studienfinanzierung und die Motivation für den gewählten Studiengang erfragt. Durch die Verknüpfung von administrativen Prüfungsdaten mit den neuen Umfragedaten wird die Analyse der Studienverläufe und des Studienerfolgs im Zusammenhang mit individuellen Merkmalen ermöglicht. Damit vermeiden wir die Schwachstellen üblicher studentischer Befragungen mit hohem Non-response und Erinnerungsfehlern. Dieser Ansatz wird hier zum Vergleich von fünf Masterstudiengängen am Fachbereich (FB) Wirtschaftswissenschaft der Freien Universität Berlin (FU Berlin) benutzt. Nach der ersten Befragung wurden die Studierenden über die ersten 6 Fachsemester hinsichtlich des Erwerbs von Leistungspunkten (LP), dem Abschluss ihres Masterstudiums sowie den dabei erzielten Noten begleitet. In den verknüpften Daten konnte ein Erfolgsbias der teilnahmebereiten Studierenden der Kohorte festgestellt werden, der aber durch eine geeignete Gewichtung über die Responserate gut beherrschbar ist. Wir vergleichen die Studiengänge in verschiedenen Phasen: Studieneingangsphase, Erreichen der Regelstudienzeit und Abschluss bis zum 6 Fachsemester. Weiterhin wird der Einfluss von Hintergrundmerkmalen auf die erreichte Note beim Studienabschluss untersucht. In der Studieneingangsphase ergeben sich deutliche Unterschiede zwischen den Studiengängen. Konditioniert man allerdings auf den Studienerfolg im ersten Semester, so verschwinden diese Unterschiede für die zweite Studienphase und den Abschluss des Studiums. Überraschend ist der geringe Einfluss der Bachelor-Note auf den Studienerfolg und die erzielte Master-Note. Die Ergebnisse zeigen die Möglichkeit auf, einen möglichen Studienabbruch schon relativ sicher anhand der erzielten LP in der Studieneingangsphase zu erkennen. KW - Administrative Prüfungsdaten KW - Umfragedaten KW - Studienerfolg KW - Master Programme KW - Studiengang Monitoring KW - Administrative Examination Data KW - Survey Data KW - Study Success KW - Master Programs KW - Study Program Monitoring ER - TY - CHAP TI - Statistische Analyse unvollständiger Daten AU - Meinfelder, Florian PY - 2023 T2 - Moderne Verfahren der Angewandten Statistik SP - 1-39 PB - Springer CY - Berlin, Heidelberg SN - 978-3-662-63496-7 DO - 10.1007/978-3-662-63496-7_24-1 UR - https://fis.uni-bamberg.de/handle/uniba/116859 LA - deu ER - TY - JOUR TI - Releasing survey microdata with exact cluster locations and additional privacy safeguards AU - Koebe, Till AU - Arias‐Salazar, Alejandra AU - Schmid, Timo PY - 2023 JO - Humanities and Social Sciences Communications VL - 10 IS - 220 SP - 1-13 PB - Springer Nature CY - London SN - 2662-9992 DO - 10.1057/s41599-023-01694-y UR - https://fis.uni-bamberg.de/handle/uniba/108611 LA - eng AB - Household survey programs around the world publish fine-granular georeferenced microdata to support research on the interdependence of human livelihoods and their surrounding environment. To safeguard the respondents’ privacy, micro-level survey data is usually (pseudo)-anonymized through deletion or perturbation procedures such as obfuscating the true location of data collection. This, however, poses a challenge to emerging approaches that augment survey data with auxiliary information on a local level. Here, we propose an alternative microdata dissemination strategy that leverages the utility of the original microdata with additional privacy safeguards through synthetically generated data using generative models. We back our proposal with experiments using data from the 2011 Costa Rican census and satellite-derived auxiliary information. Our strategy reduces the respondents’ re-identification risk for any number of disclosed attributes by 60–80% even under re-identification attempts. KW - survey microdata ER - TY - JOUR TI - Small Area with Multiply Imputed Survey Data AU - Runge, Marina AU - Schmid, Timo PY - 2023 JO - Journal of Official Statistics VL - 39 IS - 4 SP - 507-533 PB - Sage Publishing CY - Thousand Oaks, California SN - 0282-423X DO - 10.2478/jos-2023-0024 UR - https://fis.uni-bamberg.de/handle/uniba/95823 LA - eng AB - In this article, we propose a framework for small area estimation with multiply imputed survey data. Many statistical surveys suffer from (a) high nonresponse rates due to sensitive questions and response burden and (b) too small sample sizes to allow for reliable estimates on (unplanned) disaggregated levels due to budget constraints. One way to deal with missing values is to replace them by several plausible/imputed values based on a model. Small area estimation, such as the model by Fay and Herriot, is applied to estimate regionally disaggregated indicators when direct estimates are imprecise. The framework presented tackles simultaneously multiply imputed values and imprecise direct estimates. In particular, we extend the general class of transformed Fay-Herriot models to account for the additional uncertainty from multiple imputation. We derive three special cases of the Fay-Herriot model with particular transformations and provide point and mean squared error estimators. Depending on the case, the mean squared error is estimated by analytic solutions or resampling methods. Comprehensive simulations in a controlled environment show that the proposed methodology leads to reliable and precise results in terms of bias and mean squared error. The methodology is illustrated by a real data example using European wealth data. KW - Fay-Herriot model KW - mean squared error KW - multiple imputation KW - nonresponse KW - survey statistics ER - TY - JOUR TI - A Framework for Producing Small Area Estimates Based on Area-Level Models in R AU - Harmening, Sylvia AU - Kreutzmann, Ann-Kristin AU - Schmidt, Sören AU - Salvati, Nicola AU - Schmid, Timo PY - 2023 JO - The R Journal VL - 15 IS - 1 SP - 316-341 PB - The R Foundation CY - Frederiksberg SN - 2073-4859 DO - 10.32614/rj-2023-039 UR - https://fis.uni-bamberg.de/handle/uniba/92478 LA - eng AB - The R package emdi facilitates the estimation of regionally disaggregated indicators using small area estimation methods and provides tools for model building, diagnostics, presenting, and exporting the results. The package version 1.1.7 includes unit-level small area models that rely on access to micro data. The area-level model by Fay and Herriot (1979) and various extensions have been added to the package since the release of version 2.0.0. These extensions include (a) area-level models with back-transformations, (b) spatial and robust extensions, (c) adjusted variance estimation methods, and (d) area-level models that account for measurement errors. Corresponding mean squared error estimators are implemented for assessing the uncertainty. User-friendly tools like a stepwise variable selection, model diagnostics, benchmarking options, high quality maps and results exportation options enable a complete analysis procedure. The functionality of the package is illustrated by examples based on synthetic data for Austrian districts. KW - Area-Level Models in R ER - TY - JOUR TI - Editorial Heft 1, 2023: Vorwort der Herausgeber AU - Schmid, Timo AU - Zwick, Markus PY - 2023 SP - 1-3 PB - Otto-Friedrich-Universität CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/59713 LA - deu AB - - KW - Datenerfassung ER - TY - JOUR TI - Editorial Heft 1, 2023: Vorwort der Herausgeber AU - Schmid, Timo AU - Zwick, Markus PY - 2023 JO - AStA Wirtschafts- und Sozialstatistisches Archiv VL - 17 IS - 1 SP - 1-3 PB - Springer Nature CY - Berlin SN - 1863-8163 DO - 10.1007/s11943-023-00319-9 UR - https://fis.uni-bamberg.de/handle/uniba/59677 LA - deu AB - - KW - Datenerfassung ER - TY - JOUR TI - Variable selection using conditional AIC for linear mixed models with data-driven transformations AU - Lee, Yeonjoo AU - Schmid, Timo AU - Rojas‐Perilla, Natalia AU - Runge, Marina PY - 2023 SP - 1-17 PB - Otto-Friedrich-Universität CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/58402 LA - eng AB - When data analysts use linear mixed models, they usually encounter two practical problems: (a) the true model is unknown and (b) the Gaussian assumptions of the errors do not hold. While these problems commonly appear together, researchers tend to treat them individually by (a) finding an optimal model based on the conditional Akaike information criterion (cAIC) and (b) applying transformations on the dependent variable. However, the optimal model depends on the transformation and vice versa. In this paper, we aim to solve both problems simultaneously. In particular, we propose an adjusted cAIC by using the Jacobian of the particular transformation such that various model candidates with differently transformed data can be compared. From a computational perspective, we propose a step-wise selection approach based on the introduced adjusted cAIC. Model-based simulations are used to compare the proposed selection approach to alternative approaches. Finally, the introduced approach is applied to Mexican data to estimate poverty and inequality indicators for 81 municipalities. KW - Box-Cox transformation KW - Empirical best predictor KW - Indicators KW - Small area estimation ER - TY - JOUR TI - Estimating regional income indicators under transformations and access to limited population auxiliary information AU - Würz, Nora AU - Schmid, Timo AU - Tzavidis, Nikos PY - 2023 SP - 1679-1706 PB - Otto-Friedrich-Universität CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/58282 LA - eng AB - Spatially disaggregated income indicators are typically estimated by using model-based methods that assume access to auxiliary information from population micro-data. In many countries like Germany and the UK population micro-data are not publicly available. In this work we propose small area methodology when only aggregate population-level auxiliary information is available. We use data-driven transformations of the response to satisfy the parametric assumptions of the used models. In the absence of population micro-data, appropriate bias-corrections for small area prediction are needed. Under the approach we propose in this paper, aggregate statistics (means and covariances) and kernel density estimation are used to resolve the issue of not having access to population micro-data. We further explore the estimation of the mean squared error using the parametric bootstrap. Extensive model-based and design-based simulations are used to compare the proposed method to alternative methods. Finally, the proposed methodology is applied to the 2011 Socio-Economic Panel and aggregate census information from the same year to estimate the average income for 96 regional planning regions in Germany. KW - census KW - density estimation KW - official statistics KW - small area estimation KW - unit-level models ER - TY - JOUR TI - Flexible domain prediction using mixed effects random forests AU - Schmid, Timo AU - Krennmair, Patrick PY - 2023 VL - 71 IS - 5 SP - 1865-1894 PB - Otto-Friedrich-Universität CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/58270 LA - eng AB - This paper promotes the use of random forests as versatile tools for estimating spatially disaggregated indicators in the presence of small area-specific sample sizes. Small area estimators are predominantly conceptualised within the regression-setting and rely on linear mixed models to account for the hierarchical structure of the survey data. In contrast, machine learning methods offer non-linear and non-parametric alternatives, combining excellent predictive performance and a reduced risk of model-misspecification. Mixed effects random forests combine advantages of regression forests with the ability to model hierarchical dependencies. This paper provides a coherent framework based on mixed effects random forests for estimating small area averages and proposes a non-parametric bootstrap estimator for assessing the uncertainty of the estimates. We illustrate advantages of our proposed methodology using Mexican income-data from the state Nuevo León. Finally, the methodology is evaluated in model-based and design-based simulations comparing the proposed methodology to traditional regression-based approaches for estimating small area averages. KW - mean squared error KW - official statistics KW - small area estimation KW - tree-based methods ER - TY - JOUR TI - Editorial : Vorwort der Herausgeber AU - Schmid, Timo AU - Zwick, Markus PY - 2023 SP - 167–170 PB - Otto-Friedrich-Universität CY - Bamberg SN - 1863-8163 UR - https://fis.uni-bamberg.de/handle/uniba/57985 LA - deu AB - Nicht verfügbar Not available KW - Statistik KW - Wirtschaft ER - TY - JOUR TI - Statistik zwischen Data Science, Artificial Intelligence und Big Data : Beiträge aus dem Kolloquium „Make Statistics great again“ AU - Meinfelder, Florian AU - Rendtel, Ulrich AU - Seidel, Willi AU - Müller, Christine AU - Wagner, Joachim AU - Chlumsky, Jürgen AU - Zwick, Markus PY - 2023 SP - 97-147 PB - Otto-Friedrich-Universität CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/57721 LA - deu AB - Die Statistik als fachliche Disziplin muss sich in einem rasch wandelnden Umfeld behaupten, das durch den Aufstieg der Data Science, den Bedeutungszuwachs der künstlichen Intelligenz sowie neue Datenstrukturen charakterisiert wird. Wie kann sich die Statistik hier behaupten oder verlorenes Terrain wiedergewinnen? Unter dem provokanten Motto „Make Statistics great again“ wurden aus verschiedenen Blickwinkeln Entwicklungen, Strategien und positive Beispiele skizziert, wie sich das Fach Statistik an der Universität, im Wissenschaftsbetrieb und auf dem Arbeitsmarkt positionieren sollte. Willi Seidel schaut aus dem Blickwinkel eines Hochschulpräsidenten auf den Ressourcenkampf der Fächer. Christine Müller berichtet über die Initiativen der Dachorganisation DAGStat die vielen Teildisziplinen der Statistik wirkungsvoll im Wissenschaftsbetrieb und in der Öffentlichkeit zu positionieren. Florian Meinfelder dokumentiert den Aufstieg des Masterstudiengangs in Survey Statistik zu einem der nachgefragtesten Studiengänge der Uni Bamberg. Jürgen Chlumsky und Markus Zwick beleuchten die historische Wahrnehmung der Amtsstatistik bei Pflichterhebungen und die Entwicklung der Forschungsdatenzentren sowie moderner Zugänge zu neuen Datenquellen. Joachim Wagner schildert das Verhältnis von Datenproduzenten und Datennutzern aus der Sicht eines unzufriedenen Datennutzers. Schließlich geht es um die Position der Statistik in der Datenwissenschaft. Ist „Data Science“ nur ein neumodisches Wort für Statistik? Ein Konzeptionspapier der Gesellschaft für Informatik (GI) hat zu Positionspapieren der DStatG und der DAGStat geführt, die von Ulrich Rendtel vorgestellt werden. Das Kolloquium fand anlässlich der Abschiedsvorlesung von Ulrich Rendtel im Juni 2019 am Fachbereich Wirtschaftswissenschaft der Freien Universität statt. KW - Statistik Studiengänge KW - Survey Statistk KW - Amtliche Statistik KW - Datenzugang KW - Data Science KW - Statistical Literacy KW - Big Data KW - Künstliche Intelligenz KW - Informatik KW - Programmes in statistics KW - Survey statistics KW - Official statistics KW - Data access KW - Data science KW - Statistical literacy KW - Big data KW - Artificial intelligence KW - Informatics ER - TY - JOUR TI - Data Fusion for Joining Income and Consumtion Information using Different Donor-Recipient Distance Metrics AU - Meinfelder, Florian AU - Schaller, Jannik PY - 2023 SP - 509-532 PB - Otto-Friedrich-Universität CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/57719 LA - eng AB - Data fusion describes the method of combining data from (at least) two initially independent data sources to allow for joint analysis of variables which are not jointly observed. The fundamental idea is to base inference on identifying assumptions, and on common variables which provide information that is jointly observed in all the data sources. A popular class of methods dealing with this particular missing-data problem in practice is based on covariatebased nearest neighbour matching, whereas more flexible semi- or even fully parametric approaches seem underrepresented in applied data fusion. In this article we compare two different approaches of nearest neighbour hot deck matching: One, Random Hot Deck, is a variant of the covariate-based matching methods which was proposed by Eurostat, and can be considered as a ’classical’ statistical matching method, whereas the alternative approach is based on Predictive Mean Matching. We discuss results from a simulation study where we deviate from previous analyses of marginal distributions and consider joint distributions of fusion variables instead, and our findings suggest that Predictive Mean Matching tends to outperform Random Hot Deck. KW - Statistical matching KW - missing data KW - predictive mean matching KW - nearest neighbour Imputation KW - missing-by-design pattern ER - TY - JOUR TI - Variable selection using conditional AIC for linear mixed models with data-driven transformations AU - Lee, Yeonjoo AU - Schmid, Timo AU - Rojas‐Perilla, Natalia AU - Runge, Marina PY - 2023 JO - Statistics and computing VL - 33 IS - 1 SP - 1-17 PB - Springer Science + Business Media B.V CY - Dordrecht [u.a.] SN - 1573-1375 DO - 10.1007/s11222-022-10198-9 UR - https://fis.uni-bamberg.de/handle/uniba/57600 LA - eng AB - When data analysts use linear mixed models, they usually encounter two practical problems: (a) the true model is unknown and (b) the Gaussian assumptions of the errors do not hold. While these problems commonly appear together, researchers tend to treat them individually by (a) finding an optimal model based on the conditional Akaike information criterion (cAIC) and (b) applying transformations on the dependent variable. However, the optimal model depends on the transformation and vice versa. In this paper, we aim to solve both problems simultaneously. In particular, we propose an adjusted cAIC by using the Jacobian of the particular transformation such that various model candidates with differently transformed data can be compared. From a computational perspective, we propose a step-wise selection approach based on the introduced adjusted cAIC. Model-based simulations are used to compare the proposed selection approach to alternative approaches. Finally, the introduced approach is applied to Mexican data to estimate poverty and inequality indicators for 81 municipalities. KW - Box-Cox transformation KW - Empirical best predictor KW - Indicators KW - Small area estimation ER - TY - JOUR TI - Multiple imputation of ordinal missing not at random data AU - Hammon, Angelina PY - 2023 JO - Advances in statistical analysis : AStA VL - 107 IS - 4 SP - 671-692 PB - Springer CY - Berlin ; Heidelberg SN - 1863-818X DO - 10.1007/s10182-022-00461-9 UR - https://fis.uni-bamberg.de/handle/uniba/56927 LA - eng AB - We introduce a selection model-based imputation approach to be used within the Fully Conditional Specification (FCS) framework for the Multiple Imputation (MI) of incomplete ordinal variables that are supposed to be Missing Not at Random (MNAR). Thereby, we generalise previous work on this topic which involved binary single-level and multilevel data to ordinal variables. We apply an ordered probit model with sample selection as base of our imputation algorithm. The applied model involves two equations that are modelled jointly where the first one describes the missing-data mechanism and the second one specifies the variable to be imputed. In addition, we develop a version for hierarchical data by incorporating random intercept terms in both equations. To fit this multilevel imputation model we use quadrature techniques. Two simulation studies validate the overall good performance of our single-level and multilevel imputation methods. In addition, we show its applicability to empirical data by applying it to a common research topic in educational science using data of the National Educational Panel Study (NEPS) and conducting a short sensitivity analysis. Our approach is designed to be used within the R software package mice which makes it easy to access and apply. KW - Fully conditional specification KW - Missingness not at random KW - Multilevel data KW - Multiple imputation KW - Selection model KW - Ordinal data ER - TY - JOUR TI - A Bayesian Approach Towards Missing Covariate Data in Multilevel Latent Regression Models AU - Aßmann, Christian AU - Gaasch, Jean-Christoph AU - Stingl, Doris PY - 2023 JO - Psychometrika : a journal of quantitative psychology VL - 88 IS - 4 SP - 1495-1528 PB - Springer CY - New York SN - 1860-0980 DO - 10.1007/s11336-022-09888-0 UR - https://fis.uni-bamberg.de/handle/uniba/56766 LA - eng AB - The measurement of latent traits and investigation of relations between these and a potentially large set of explaining variables is typical in psychology, economics, and the social sciences. Corresponding analysis often relies on surveyed data from large-scale studies involving hierarchical structures and missing values in the set of considered covariates. This paper proposes a Bayesian estimation approach based on the device of data augmentation that addresses the handling of missing values in multilevel latent regression models. Population heterogeneity is modeled via multiple groups enriched with random intercepts. Bayesian estimation is implemented in terms of a Markov chain Monte Carlo sampling approach. To handle missing values, the sampling scheme is augmented to incorporate sampling from the full conditional distributions of missing values. We suggest to model the full conditional distributions of missing values in terms of non-parametric classification and regression trees. This offers the possibility to consider information from latent quantities functioning as sufficient statistics. A simulation study reveals that this Bayesian approach provides valid inference and outperforms complete cases analysis and multiple imputation in terms of statistical efficiency and computation time involved. An empirical illustration using data on mathematical competencies demonstrates the usefulness of the suggested approach. KW - Item response theory KW - population heterogeneity KW - Markov chain Monte Carlo KW - classification and regression trees KW - missing values ER - TY - JOUR TI - Editorial : Vorwort der Herausgeber AU - Schmid, Timo AU - Zwick, Markus PY - 2022 JO - AStA Wirtschafts- und Sozialstatistisches Archiv VL - 16 IS - 3-4 SP - 167-170 PB - Springer CY - Berlin ; Heidelberg SN - 1863-8163 DO - 10.1007/s11943-022-00315-5 UR - https://fis.uni-bamberg.de/handle/uniba/57984 LA - deu AB - Nicht verfügbar Not available KW - Statistik KW - Wirtschaft ER - TY - JOUR TI - Vorwort der Herausgeber AU - Schmid, Timo AU - Zwick, Markus PY - 2022 SP - 1-4 PB - Otto-Friedrich-Universität CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/55976 LA - deu AB - Nicht verfügbar Not available KW - Grundschulversorgung KW - COVID-19-Pandemie KW - Berliner Wahldaten KW - Übersterblichkeit KW - Räumliche Daten KW - Deutschen Arbeitsmarkt ER - TY - JOUR TI - Statistik zwischen Data Science, Artificial Intelligence und Big Data : Beiträge aus dem Kolloquium „Make Statistics great again“ AU - Meinfelder, Florian AU - Rendtel, Ulrich AU - Seidel, Willi AU - Müller, Christine AU - Wagner, Joachim AU - Chlumsky, Jürgen AU - Zwick, Markus PY - 2022 JO - Wirtschafts- und sozialstatistisches Archiv : eine Zeitschrift der Deutschen Statistischen Gesellschaft VL - 16 IS - 2 SP - 97-147 PB - Springer CY - Berlin ; Heidelberg SN - 1863-8163 DO - 10.1007/s11943-022-00305-7 UR - https://fis.uni-bamberg.de/handle/uniba/57546 LA - deu AB - Die Statistik als fachliche Disziplin muss sich in einem rasch wandelnden Umfeld behaupten, das durch den Aufstieg der Data Science, den Bedeutungszuwachs der künstlichen Intelligenz sowie neue Datenstrukturen charakterisiert wird. Wie kann sich die Statistik hier behaupten oder verlorenes Terrain wiedergewinnen? Unter dem provokanten Motto „Make Statistics great again“ wurden aus verschiedenen Blickwinkeln Entwicklungen, Strategien und positive Beispiele skizziert, wie sich das Fach Statistik an der Universität, im Wissenschaftsbetrieb und auf dem Arbeitsmarkt positionieren sollte. Willi Seidel schaut aus dem Blickwinkel eines Hochschulpräsidenten auf den Ressourcenkampf der Fächer. Christine Müller berichtet über die Initiativen der Dachorganisation DAGStat die vielen Teildisziplinen der Statistik wirkungsvoll im Wissenschaftsbetrieb und in der Öffentlichkeit zu positionieren. Florian Meinfelder dokumentiert den Aufstieg des Masterstudiengangs in Survey Statistik zu einem der nachgefragtesten Studiengänge der Uni Bamberg. Jürgen Chlumsky und Markus Zwick beleuchten die historische Wahrnehmung der Amtsstatistik bei Pflichterhebungen und die Entwicklung der Forschungsdatenzentren sowie moderner Zugänge zu neuen Datenquellen. Joachim Wagner schildert das Verhältnis von Datenproduzenten und Datennutzern aus der Sicht eines unzufriedenen Datennutzers. Schließlich geht es um die Position der Statistik in der Datenwissenschaft. Ist „Data Science“ nur ein neumodisches Wort für Statistik? Ein Konzeptionspapier der Gesellschaft für Informatik (GI) hat zu Positionspapieren der DStatG und der DAGStat geführt, die von Ulrich Rendtel vorgestellt werden. Das Kolloquium fand anlässlich der Abschiedsvorlesung von Ulrich Rendtel im Juni 2019 am Fachbereich Wirtschaftswissenschaft der Freien Universität statt. KW - Statistik Studiengänge KW - Survey Statistk KW - Amtliche Statistik KW - Datenzugang KW - Data Science KW - Statistical Literacy KW - Big Data KW - Künstliche Intelligenz KW - Informatik KW - Programmes in statistics KW - Survey statistics KW - Official statistics KW - Data access KW - Data science KW - Statistical literacy KW - Big data KW - Artificial intelligence KW - Informatics ER - TY - JOUR TI - Data Fusion for Joining Income and Consumtion Information using Different Donor-Recipient Distance Metrics AU - Meinfelder, Florian AU - Schaller, Jannik PY - 2022 JO - Journal of official statistics : JOS VL - 38 IS - 2 SP - 509-532 PB - de Gruyter CY - Berlin SN - 2001-7367 DO - 10.2478/jos-2022-0024 UR - https://fis.uni-bamberg.de/handle/uniba/57545 LA - eng AB - Data fusion describes the method of combining data from (at least) two initially independent data sources to allow for joint analysis of variables which are not jointly observed. The fundamental idea is to base inference on identifying assumptions, and on common variables which provide information that is jointly observed in all the data sources. A popular class of methods dealing with this particular missing-data problem in practice is based on covariatebased nearest neighbour matching, whereas more flexible semi- or even fully parametric approaches seem underrepresented in applied data fusion. In this article we compare two different approaches of nearest neighbour hot deck matching: One, Random Hot Deck, is a variant of the covariate-based matching methods which was proposed by Eurostat, and can be considered as a ’classical’ statistical matching method, whereas the alternative approach is based on Predictive Mean Matching. We discuss results from a simulation study where we deviate from previous analyses of marginal distributions and consider joint distributions of fusion variables instead, and our findings suggest that Predictive Mean Matching tends to outperform Random Hot Deck. KW - Statistical matching KW - missing data KW - predictive mean matching KW - nearest neighbour Imputation KW - missing-by-design pattern ER - TY - JOUR TI - Flexible domain prediction using mixed effects random forests AU - Schmid, Timo AU - Krennmair, Patrick PY - 2022 JO - Journal of the Royal Statistical Society. Series C, Applied statistics VL - 71 IS - 5 SP - 1865-1894 PB - Wiley-Blackwell CY - London SN - 1467-9876 DO - 10.1111/rssc.12600 UR - https://fis.uni-bamberg.de/handle/uniba/57459 LA - eng AB - This paper promotes the use of random forests as versatile tools for estimating spatially disaggregated indicators in the presence of small area-specific sample sizes. Small area estimators are predominantly conceptualised within the regression-setting and rely on linear mixed models to account for the hierarchical structure of the survey data. In contrast, machine learning methods offer non-linear and non-parametric alternatives, combining excellent predictive performance and a reduced risk of model-misspecification. Mixed effects random forests combine advantages of regression forests with the ability to model hierarchical dependencies. This paper provides a coherent framework based on mixed effects random forests for estimating small area averages and proposes a non-parametric bootstrap estimator for assessing the uncertainty of the estimates. We illustrate advantages of our proposed methodology using Mexican income-data from the state Nuevo León. Finally, the methodology is evaluated in model-based and design-based simulations comparing the proposed methodology to traditional regression-based approaches for estimating small area averages. KW - mean squared error KW - official statistics KW - small area estimation KW - tree-based methods ER - TY - JOUR TI - Estimating regional income indicators under transformations and access to limited population auxiliary information AU - Würz, Nora AU - Schmid, Timo AU - Tzavidis, Nikos PY - 2022 JO - Journal of the Royal Statistical Society. Series A, Statistics in society VL - 185 IS - 4 SP - 1679-1706 PB - Wiley-Blackwell CY - London SN - 1467-985X DO - 10.1111/rssa.12913 UR - https://fis.uni-bamberg.de/handle/uniba/57458 LA - eng AB - Spatially disaggregated income indicators are typically estimated by using model-based methods that assume access to auxiliary information from population micro-data. In many countries like Germany and the UK population micro-data are not publicly available. In this work we propose small area methodology when only aggregate population-level auxiliary information is available. We use data-driven transformations of the response to satisfy the parametric assumptions of the used models. In the absence of population micro-data, appropriate bias-corrections for small area prediction are needed. Under the approach we propose in this paper, aggregate statistics (means and covariances) and kernel density estimation are used to resolve the issue of not having access to population micro-data. We further explore the estimation of the mean squared error using the parametric bootstrap. Extensive model-based and design-based simulations are used to compare the proposed method to alternative methods. Finally, the proposed methodology is applied to the 2011 Socio-Economic Panel and aggregate census information from the same year to estimate the average income for 96 regional planning regions in Germany. KW - census KW - density estimation KW - official statistics KW - small area estimation KW - unit-level models ER - TY - JOUR TI - Iterative Kernel Density Estimation Applied to Grouped Data : Estimating Poverty and Inequality Indicators from the German Microcensus AU - Walter, Paul AU - Groß, Marcus AU - Schmid, Timo AU - Weimer, Katja PY - 2022 SP - 599-635 PB - Otto-Friedrich-Universität CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/55983 LA - eng AB - The estimation of poverty and inequality indicators based on survey data is trivial as long as the variable of interest (e.g., income or consumption) is measured on a metric scale. However, estimation is not directly possible, using standard formulas, when the income variable is grouped due to confidentiality constraints or in order to decrease item nonresponse. We propose an iterative kernel density algorithm that generates metric pseudo samples from the grouped variable for the estimation of indicators. The corresponding standard errors are estimated by a non-parametric bootstrap that accounts for the additional uncertainty due to the grouping. The algorithm enables the use of survey weights and household equivalence scales. The proposed method is applied to the German Microcensus for estimating the regional distribution of poverty and inequality in Germany. KW - Direct estimation KW - Interval-censored data KW - non-parametric estimation KW - OECD scale KW - prediction ER - TY - JOUR TI - Vorwort der Herausgeber AU - Schmid, Timo AU - Zwick, Markus PY - 2022 JO - AStA Wirtschafts- und Sozialstatistisches Archiv VL - 16 IS - 1 SP - 1-4 PB - Springer CY - Berlin ; Heidelberg SN - 1863-8163 DO - 10.1007/s11943-022-00304-8 UR - https://fis.uni-bamberg.de/handle/uniba/55975 LA - deu AB - Nicht verfügbar Not available KW - Grundschulversorgung KW - COVID-19-Pandemie KW - Berliner Wahldaten KW - Übersterblichkeit KW - Räumliche Daten KW - Deutschen Arbeitsmarkt ER - TY - JOUR TI - Iterative Kernel Density Estimation Applied to Grouped Data : Estimating Poverty and Inequality Indicators from the German Microcensus AU - Walter, Paul AU - Groß, Marcus AU - Schmid, Timo AU - Weimer, Katja PY - 2022 JO - Journal of Official Statistics : JOS VL - 38 IS - 2 SP - 599-635 PB - de Gruyter CY - Berlin SN - 2001-7367 DO - 10.2478/jos-2022-0027 UR - https://fis.uni-bamberg.de/handle/uniba/55652 LA - eng AB - The estimation of poverty and inequality indicators based on survey data is trivial as long as the variable of interest (e.g., income or consumption) is measured on a metric scale. However, estimation is not directly possible, using standard formulas, when the income variable is grouped due to confidentiality constraints or in order to decrease item nonresponse. We propose an iterative kernel density algorithm that generates metric pseudo samples from the grouped variable for the estimation of indicators. The corresponding standard errors are estimated by a non-parametric bootstrap that accounts for the additional uncertainty due to the grouping. The algorithm enables the use of survey weights and household equivalence scales. The proposed method is applied to the German Microcensus for estimating the regional distribution of poverty and inequality in Germany. KW - Direct estimation KW - Interval-censored data KW - non-parametric estimation KW - OECD scale KW - prediction ER - TY - JOUR TI - Bayesian estimation and model comparison for linear dynamic panel models with missing values AU - Aßmann, Christian AU - Preising, Marcel PY - 2022 SP - 536-557 PB - Otto-Friedrich-Universität CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/54962 LA - eng AB - Panel data are collected over several time periods for the same units and hence allow for modelling both latent heterogeneity and dynamics. Since in a dynamic setup, the dependent variable also appears as an explanatory variable in later periods, missing values lead to substantial loss of information and the possibility of inefficient estimation. For linear dynamic panel models with fixed or random effects, we suggest a Bayesian approach to deal with missing values. The Gibbs sampling scheme providing a sample from the posterior distribution is thereby augmented by draws from the full conditional distribution of the missing values. While the full conditional distribution for missing values in the dependent variable is implied by the model setup, we incorporate a flexible non-parametric approximation to the full conditional posterior distribution of missing values in the explaining variables. Also, we provide accurate non-nested model comparison in terms of the marginal likelihood from the resulting hybrid Gibbs sampling output. The properties and possible efficiency gains of the suggested approach are illustrated by means of a simulation study and an empirical application using a macroeconomic panel data set. KW - data augmentation KW - dynamic linear panel regression KW - marginal likelihood KW - missing values ER - TY - JOUR TI - Domain prediction with grouped income data AU - Walter, Paul AU - Groß, Marcus AU - Schmid, Timo AU - Tzavidis, Nikos PY - 2022 SP - 1501-1523 PB - Otto-Friedrich-Universität CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/54774 LA - eng AB - One popular small area estimation method for estimating poverty and inequality indicators is the empirical best predictor under the unit-level nested error regression model with a continuous dependent variable. However, parameter estimation is more challenging when the response variable is grouped due to data confidentiality concerns or concerns about survey response burden. The work in this paper proposes methodology that enables fitting a nested error regression model when the dependent variable is grouped. Model parameters are then used for small area prediction of finite population parameters of interest. Model fitting in the case of a grouped response variable is based on the use of a stochastic expectation–maximization algorithm. Since the stochastic expectation–maximization algorithm relies on the Gaussian assumptions of the unit-level error terms, adaptive transformations are incorporated for handling departures from normality. The estimation of the mean squared error of the small area parameters is facilitated by a parametric bootstrap that captures the additional uncertainty due to the grouping mechanism and the possible use of adaptive transformations. The empirical properties of the proposed methodology are assessed by using model-based simulations and its relevance is illustrated by estimating deprivation indicators for municipalities in the Mexican state of Chiapas. KW - data confidentiality KW - interval-censored data KW - nested error regression model KW - small area estimation KW - survey response burden ER - TY - JOUR TI - Experimental UK Regional Consumer Price Inflation with Model-Based Expenditure Weights AU - Dawber, James AU - Würz, Nora AU - Smith, Paul A. AU - Flower, Tanya AU - Thomas, Heledd AU - Schmid, Timo AU - Tzavidis, Nikos PY - 2022 JO - Journal of Official Statistics VL - 38 IS - 1 SP - 213-237 PB - de Gruyter CY - Berlin SN - 2001-7367 DO - 10.2478/jos-2022-0010 UR - https://fis.uni-bamberg.de/handle/uniba/53876 LA - deu AB - Like many other countries, the United Kingdom (UK) produces a national consumer price index (CPI) to measure inflation. Presently, CPI measures are not produced for regions within the UK. It is believed that, using only available data sources, a regional CPI would not be precise or reliable enough as an official statistic, primarily because the regional partitioning of the data makes sample sizes too small. We investigate this claim by producing experimental regional CPIs using publicly available price data, and deriving expenditure weights from the Living Costs and Food survey. We detail the methods and challenges of developing a regional CPI and evaluate its reliability. We then assess whether model-based methods such as smoothing and small area estimation significantly improve the measures. We find that a regional CPI can be produced with available data sources, however it appears to be excessively volatile over time, mainly due to the weights. Smoothing and small area estimation improve the reliability of the regional CPI series to some extent but they remain too volatile for regional policy use. This research provides a valuable framework for the development of a more viable regional CPI measure for the UK in the future. KW - CPI conceptual framework KW - basket of goods and services KW - small area estimation KW - Fay-Herriot models ER - TY - JOUR TI - Intercensal updating using structure‐preserving methods and satellite imagery AU - Koebe, Till AU - Arias‐Salazar, Alejandra AU - Rojas‐Perilla, Natalia AU - Schmid, Timo PY - 2022 JO - Journal of the Royal Statistical Society : Series A, Statistics in society VL - 185 IS - Supplement 2 SP - 1-27 PB - Oxford University Press CY - London SN - 1467-985X DO - 10.1111/rssa.12802 UR - https://fis.uni-bamberg.de/handle/uniba/53875 LA - eng AB - Censuses are fundamental building blocks of most modern-day societies, yet collected every 10 years at best. We propose an extension of the widely popular census updating technique structure-preserving estimation by incorporating auxiliary information in order to take ongoing subnational population shifts into account. We apply our method by incorporating satellite imagery as additional source to derive annual small-area updates of multidimensional poverty indicators from 2013 to 2020 for a population at risk: female-headed households in Senegal. We evaluate the performance of our proposal using data from two different census periods. KW - multidimensional poverty KW - official statistics KW - small-area estimation KW - SPREE ER - TY - JOUR TI - Editorial AU - Schmid, Timo AU - Zwick, Markus PY - 2021 JO - Wirtschafts- und sozialstatistisches Archiv : ASTA ; eine Zeitschrift der Deutschen Statistischen Gesellschaft VL - 15 IS - 2 SP - 59-61 PB - Springer CY - Berlin ; Heidelberg SN - 1863-8155 DO - 10.1007/s11943-021-00291-2 UR - https://fis.uni-bamberg.de/handle/uniba/105832 LA - deu AB - - KW - - ER - TY - JOUR TI - Editorial AU - Zwick, Markus AU - Schmid, Timo PY - 2021 JO - Wirtschafts- und sozialstatistisches Archiv : ASTA ; eine Zeitschrift der Deutschen Statistischen Gesellschaft VL - 15 IS - 1 SP - 1-3 PB - Springer CY - Berlin ; Heidelberg SN - 1863-8155 DO - 10.1007/s11943-021-00283-2 UR - https://fis.uni-bamberg.de/handle/uniba/105831 LA - deu AB - - KW - - ER - TY - JOUR TI - Kernel density smoothing of composite spatial data on administrative area level AU - Erfurth, Kerstin AU - Groß, Marcus AU - Rendtel, Ulrich AU - Schmid, Timo PY - 2021 JO - AStA Wirtschafts- und Sozialstatistisches Archiv VL - 16 IS - 1 SP - 25-49 PB - Heidelberg : Springer CY - Berlin SN - 1863-8163 DO - 10.1007/s11943-021-00298-9 UR - https://fis.uni-bamberg.de/handle/uniba/53874 LA - eng AB - Composite spatial data on administrative area level are often presented by maps. The aim is to detect regional differences in the concentration of subpopulations, like elderly persons, ethnic minorities, low-educated persons, voters of a political party or persons with a certain disease. Thematic collections of such maps are presented in different atlases. The standard presentation is by Choropleth maps where each administrative unit is represented by a single value. These maps can be criticized under three aspects: the implicit assumption of a uniform distribution within the area, the instability of the resulting map with respect to a change of the reference area and the discontinuities of the maps at the borderlines of the reference areas which inhibit the detection of regional clusters. In order to address these problems we use a density approach in the construction of maps. This approach does not enforce a local uniform distribution. It does not depend on a specifc choice of area reference system and there are no discontinuities in the displayed maps. A standard estimation procedure of densities are Kernel density estimates. However, these estimates need the geo-coordinates of the single units which are not at disposal as we have only access to the aggregates of some area system. To overcome this hurdle, we use a statistical simulation concept. This can be interpreted as a Simulated Expectation Maximisation (SEM) algorithm of Celeux et al (1996). We simulate observations from the current density estimates which are consistent with the aggregation information (S-step). Then we apply the Kernel density estimator to the simulated sample which gives the next density estimate (E-Step). This concept has been f rst applied for grid data with rectangular areas, see Groß et al (2017), for the display of ethnic minorities. In a second application we demonstrated the use of this approach for the so-called “change of support” (Bradley et al 2016) problem. Here Groß et al (2020) used the SEM algorithm to recalculate case numbers between non-hierarchical administrative area systems. Recently Rendtel et al (2021) applied the SEM algorithm to display spatial-temporal clusters of Corona infections in Germany. Here we present three modif cations of the basic SEM algorithm: 1) We introduce a boundary correction which removes the underestimation of kernel density estimates at the borders of the population area. 2) We recognize unsettled areas, like lakes, parks and industrial areas, in the computation of the kernel density. 3) We adapt the SEM algorithm for the computation of local percentages which are important especially in voting analysis. We evaluate our approach against several standard maps by means of the local voting register with known addresses. In the empirical part we apply our approach for the display of voting results for the 2016 election of the Berlin parliament. We contrast our results against Choropleth maps and show new possibilities for reporting spatial voting results. KW - Spatial data KW - Administrative areas KW - Choropleths KW - Kernel density estimation KW - Voting atlases ER - TY - JOUR TI - The Fay–Herriot model for multiply imputed data with an application to regional wealth estimation in Germany AU - Kreutzmann, Ann-Kristin AU - Marek, Philipp AU - Runge, Marina AU - Salvati, Nicola AU - Schmid, Timo PY - 2021 JO - Journal of applied statistics VL - 49 IS - 13 SP - 3278-3299 PB - Taylor & Francis CY - Abingdon [u.a.] SN - 0266-4763 DO - 10.1080/02664763.2021.1941805 UR - https://fis.uni-bamberg.de/handle/uniba/52219 LA - eng ER - TY - JOUR TI - Domain prediction with grouped income data AU - Walter, Paul AU - Groß, Marcus AU - Schmid, Timo AU - Tzavidis, Nikos PY - 2021 JO - Journal of the Royal Statistical Society. Series A, Statistics in society VL - 184 IS - 4 SP - 1501-1523 PB - Wiley-Blackwell CY - London SN - 0964-1998 DO - 10.1111/rssa.12736 UR - https://fis.uni-bamberg.de/handle/uniba/52207 LA - eng AB - One popular small area estimation method for estimating poverty and inequality indicators is the empirical best predictor under the unit-level nested error regression model with a continuous dependent variable. However, parameter estimation is more challenging when the response variable is grouped due to data confidentiality concerns or concerns about survey response burden. The work in this paper proposes methodology that enables fitting a nested error regression model when the dependent variable is grouped. Model parameters are then used for small area prediction of finite population parameters of interest. Model fitting in the case of a grouped response variable is based on the use of a stochastic expectation–maximization algorithm. Since the stochastic expectation–maximization algorithm relies on the Gaussian assumptions of the unit-level error terms, adaptive transformations are incorporated for handling departures from normality. The estimation of the mean squared error of the small area parameters is facilitated by a parametric bootstrap that captures the additional uncertainty due to the grouping mechanism and the possible use of adaptive transformations. The empirical properties of the proposed methodology are assessed by using model-based simulations and its relevance is illustrated by estimating deprivation indicators for municipalities in the Mexican state of Chiapas. KW - data confidentiality KW - interval-censored data KW - nested error regression model KW - small area estimation KW - survey response burden ER - TY - JOUR TI - Multiple imputation of binary multilevel missing not at random data AU - Hammon, Angelina AU - Zinn, Sabine PY - 2021 SP - 547-564 PB - Otto-Friedrich-Universität CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/49823 LA - eng AB - We introduce a selection model-based multilevel imputation approach to be used within the fully conditional specification framework for multiple imputation. Concretely, we apply a censored bivariate probit model to describe binary variables assumed to be missing not at random. The first equation of the model defines the regression model for the missing data mechanism. The second equation specifies the regression model of the variable to be imputed. The non-random selection of the binary data is mapped by correlations between the error terms of the two regression models. Hierarchical data structures are modelled by random intercepts in both equations. To fit the novel imputation model we use maximum likelihood and adaptive Gauss–Hermite quadrature. A comprehensive simulation study shows the overall performance of the approach.We test its usefulness for empirical research by applying it to a common problem in social scientific research: the emergence of educational aspirations. Our software is designed to be used in the R package mice. KW - Fully conditional specification; Missingness not at random; Multilevel data; Multiple imputation; Selection model ER - TY - JOUR TI - Multiple imputation of binary multilevel missing not at random data AU - Hammon, Angelina AU - Zinn, Sabine PY - 2020 JO - Journal of the Royal Statistical Society: Series C (Applied Statistics) VL - 69 IS - 3 SP - 547-564 PB - Wiley-Blackwell CY - Oxford SN - 1467-9876 DO - 10.1111/rssc.12401 UR - https://fis.uni-bamberg.de/handle/uniba/54235 LA - eng AB - We introduce a selection model-based multilevel imputation approach to be used within the fully conditional specification framework for multiple imputation. Concretely, we apply a censored bivariate probit model to describe binary variables assumed to be missing not at random. The first equation of the model defines the regression model for the missing data mechanism. The second equation specifies the regression model of the variable to be imputed. The non-random selection of the binary data is mapped by correlations between the error terms of the two regression models. Hierarchical data structures are modelled by random intercepts in both equations. To fit the novel imputation model we use maximum likelihood and adaptive Gauss–Hermite quadrature. A comprehensive simulation study shows the overall performance of the approach.We test its usefulness for empirical research by applying it to a common problem in social scientific research: the emergence of educational aspirations. Our software is designed to be used in the R package mice. KW - Fully conditional specification; Missingness not at random; Multilevel data; Multiple imputation; Selection model ER - TY - JOUR TI - Bayesian estimation and model comparison for linear dynamic panel models with missing values AU - Aßmann, Christian AU - Preising, Marcel PY - 2020 JO - Australian & New Zealand Journal of Statistics VL - 62 IS - 4 SP - 536–557 PB - Wiley-Blackwell CY - Oxford SN - 1369-1473 DO - 10.1111/anzs.12316 UR - https://fis.uni-bamberg.de/handle/uniba/49553 LA - eng AB - Panel data are collected over several time periods for the same units and hence allow for modelling both latent heterogeneity and dynamics. Since in a dynamic setup, the dependent variable also appears as an explanatory variable in later periods, missing values lead to substantial loss of information and the possibility of inefficient estimation. For linear dynamic panel models with fixed or random effects, we suggest a Bayesian approach to deal with missing values. The Gibbs sampling scheme providing a sample from the posterior distribution is thereby augmented by draws from the full conditional distribution of the missing values. While the full conditional distribution for missing values in the dependent variable is implied by the model setup, we incorporate a flexible non-parametric approximation to the full conditional posterior distribution of missing values in the explaining variables. Also, we provide accurate non-nested model comparison in terms of the marginal likelihood from the resulting hybrid Gibbs sampling output. The properties and possible efficiency gains of the suggested approach are illustrated by means of a simulation study and an empirical application using a macroeconomic panel data set. KW - data augmentation KW - dynamic linear panel regression KW - marginal likelihood KW - missing values ER - TY - JOUR TI - The R Package emdi for Estimating and Mapping Regionally Disaggregated Indicators AU - Kreutzmann, Ann-Kristin AU - Pannier, Sörenm AU - Rojas‐Perilla, Natalia AU - Schmid, Timo AU - Templ, Matthias AU - Tzavidis, Nikos PY - 2019 JO - Journal of statistical software VL - 91 IS - 7 SP - 1–33 CY - Los Angeles, Calif. SN - 1548-7660 DO - 10.18637/jss.v091.i07 UR - https://fis.uni-bamberg.de/handle/uniba/52224 LA - eng ER - TY - JOUR TI - In Memory of Professor Susanne Rässler AU - Drechsler, Jörg AU - Kiesl, Hans AU - Meinfelder, Florian AU - Raghunathan, Trivellore E. AU - Rubin, Donald B. AU - Schenker, Nathaniel AU - Zell, Elizabeth R. PY - 2019 JO - Journal of Official Statistics VL - 35 IS - 1 SP - 285-286 PB - de Gruyter CY - Berlin SN - 2001-7367 DO - 10.2478/jos-2019-0013 UR - https://fis.uni-bamberg.de/handle/uniba/50340 LA - eng ER - TY - JOUR TI - Estimation within the new integrated system of household surveys in Germany AU - Kamgar, Saeideh AU - Meinfelder, Florian AU - Münnich, Ralf AU - Navvabpour, Hamidreza PY - 2018 JO - Statistical Papers VL - 61 IS - 5 SP - 2091-2117 PB - Springer CY - Berlin ; Heidelberg [u.a.] SN - 0932-5026 DO - 10.1007/s00362-018-1023-z UR - https://fis.uni-bamberg.de/handle/uniba/50342 LA - eng ER - TY - CPAPER TI - Towards Multiple-Imputation-Proper Predictive Mean Matching AU - Gaffert, Philipp AU - Meinfelder, Florian AU - Bosch, Volker PY - 2018 IS - Survery Research Methods Section CY - Alexandria, VA UR - https://fis.uni-bamberg.de/handle/uniba/50341 LA - eng ER - TY - THES TI - Missing values in linear dynamic panel models : a Bayesian approach towards model comparison AU - Preising, Marcel PY - 2018 CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/42943 LA - eng AB - Longitudinal data are collected over several time periods for the same units and therefore allow for modeling both latent heterogeneity and dynamics. Unfortunately, the analyst is often confronted with serious estimation problems that arise due to missing data. As the dependent variable within dynamic setups additionally serves as explaining variable in later periods, missing values cause an even higher loss of information and potential to inefficient estimation than in static panel models. For linear dynamic panel models with fixed or random effects, the thesis suggests a Bayesian approach to deal with missing values. Therefore, a Gibbs sampling scheme providing a sample from the model parameter posterior distribution is augmented by draws from the full conditional distribution of the missing values. While the full conditional distribution for missing values in the dependent variable is implied by the model setup, a flexible non-parametric approximation of the full conditional posterior distribution of missing values in the explaining variables is incorporated. The thesis also provides non-nested model comparison in terms of the marginal likelihood from the resulting Gibbs sampling output. The properties and efficiency gains of the suggested estimation approach are evaluated by means of a simulation study. We compare the biases, coverage rates, root mean squared errors and model selections resulting from the data analyses before deletion and with missing values. Two empirical applications, the first modeling human growth, and the second investigating current account balances, illustrate the wide field of application of the developed approach. ER - TY - JOUR TI - Necessary, sufficient or redundant? : On the role of Country-specific legal institutions and firm-specific intangible assets for the valuation impact of corporate multinationality AU - Eckert, Stefan AU - Dittfeld, Marcus AU - Meinfelder, Florian PY - 2016 JO - European Journal of Management VL - 16 IS - 1 SP - 23-42 CY - Stockton, Ca, USA SN - 1555-4015 DO - 10.18374/EJM-16-1.3 UR - https://fis.uni-bamberg.de/handle/uniba/50344 LA - eng ER - TY - JOUR TI - Wary Partners : Strategic Portfolio Allocation and Coalition Governance in Parliamentary Democracies AU - Fernandes, Jorge M. AU - Meinfelder, Florian AU - Moury, Catherine PY - 2016 JO - Comparative political studies : CPS VL - 49 IS - 9 SP - 1270-1300 PB - Sage CY - Thousand Oaks, Calif. [u.a.] SN - 0010-4140 DO - 10.1177/0010414016628189 UR - https://fis.uni-bamberg.de/handle/uniba/50343 LA - eng ER -