TY - JOUR TI - Avoiding “Too Tall” and “Too Short” : The Effect of the Community on the Regulation of Body Height AU - Hermanussen, Michael AU - Aßmann, Christian AU - Scheffler, Christiane PY - 2026 SP - 1-14 PB - Otto-Friedrich-Universität CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/116658 LA - eng AB - Background: Recent evidence emphasizes the role of social signaling in the regulation of human growth. Aim: To disentangle the influence of physical living conditions such as wealth, health, nutrition, and education from influences that are transmitted among members of the social group. We consider the spectrum of all historically possible heights (“transgenerational growth potential”) and disentangle the influence of the physical living conditions from influences that are transmitted among members of the same social group. We ask (1) what is the magnitude of the “transgenerational growth potential”? and (2) to what extent narrows this potential upon entering a specific historic community? Samples and Methods: We report on the height of more than 14 million German conscripts and recruits born between 1865 and 1975. Results: Between the late 19th and the late 20th centuries, mean male height increased from 165.8 cm (SD 6.5 cm) to 180.1 cm (SD 7.0 cm). The skewness of the height distribution was always close to zero. Height was statistically associated with living conditions, but the association disappeared when linking characteristics of the within population distribution of height with the disparity of living conditions and economic inequality. Mean height of the social community is a strong attractor of individual height and reduces the “transgenerational growth potential” by more than 50%. Conclusion: Strong effects of the social community outweigh the effect of individual living conditions and substantially narrow the transgenerational growth potential, to protect against being “too tall” or “too short” within the social community. KW - community effect on height KW - education KW - Gini coefficient KW - gross domestic product KW - infant mortality ER - TY - JOUR TI - Avoiding “Too Tall” and “Too Short” : The Effect of the Community on the Regulation of Body Height AU - Hermanussen, Michael AU - Aßmann, Christian AU - Scheffler, Christiane PY - 2025 JO - American Journal of Human Biology VL - 37 IS - 6, e70085 SP - 1-14 PB - Wiley Interscience CY - New York, NY [u.a.] SN - 1520-6300 DO - 10.1002/ajhb.70085 UR - https://fis.uni-bamberg.de/handle/uniba/116657 LA - eng AB - Background: Recent evidence emphasizes the role of social signaling in the regulation of human growth. Aim: To disentangle the influence of physical living conditions such as wealth, health, nutrition, and education from influences that are transmitted among members of the social group. We consider the spectrum of all historically possible heights (“transgenerational growth potential”) and disentangle the influence of the physical living conditions from influences that are transmitted among members of the same social group. We ask (1) what is the magnitude of the “transgenerational growth potential”? and (2) to what extent narrows this potential upon entering a specific historic community? Samples and Methods: We report on the height of more than 14 million German conscripts and recruits born between 1865 and 1975. Results: Between the late 19th and the late 20th centuries, mean male height increased from 165.8 cm (SD 6.5 cm) to 180.1 cm (SD 7.0 cm). The skewness of the height distribution was always close to zero. Height was statistically associated with living conditions, but the association disappeared when linking characteristics of the within population distribution of height with the disparity of living conditions and economic inequality. Mean height of the social community is a strong attractor of individual height and reduces the “transgenerational growth potential” by more than 50%. Conclusion: Strong effects of the social community outweigh the effect of individual living conditions and substantially narrow the transgenerational growth potential, to protect against being “too tall” or “too short” within the social community. KW - community effect on height KW - education KW - Gini coefficient KW - gross domestic product KW - infant mortality ER - TY - JOUR TI - Sex, Sport, IGF-1 and the community effect in height hypothesis AU - Aßmann, Christian AU - Hermanussen, Michael AU - Blum, Werner F. AU - Bogin, Barry PY - 2025 SP - 4816-4832 PB - Otto-Friedrich-Universität CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/110800 LA - eng AB - We test the hypothesis that differences in social status between groups of people within a population may induce variation in insulin-like growth factor-1(IGF-1) levels and, by extension, growth in height. This is called the community effect in height hypothesis. The relationship between IGF-1, assessed via finger-prick dried blood spot, and elite level sport competition outcomes were analysed for a sample of 116 undergraduate men and women. There was a statistically significant difference between winners and losers of a competition. Winners, as a group, had higher average pre-game and post-game IGF-1 levels than losers. We proposed this type of difference as a proxy for social dominance. We found no evidence that winners increased in IGF-1 levels over losers or that members of the same team were more similar in IGF-1 levels than they were to players from other teams. These findings provide limited support toward the community effect in height hypothesis. The findings are discussed in relation to the action of the growth hormone/IGF-1 axis as a transducer of multiple bio-social influences into a coherent signal which allows the growing human to adjust and adapt to local ecological conditions. KW - GH/IGF-1 KW - social dominance KW - social networks ER - TY - JOUR TI - Height and skeletal morphology in relation to modern life style AU - Hermanussen, Michael AU - Scheffler, Christiane AU - Groth, Detlef AU - Aßmann, Christian PY - 2025 SP - 1-5 PB - Otto-Friedrich-Universität CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/109759 LA - eng AB - Height and skeletal morphology strongly relate to life style. Parallel to the decrease in physical activity and locomotion, modern people are slimmer in skeletal proportions. In German children and adolescents, elbow breadth and particularly relative pelvic breadth (50th centile of bicristal distance divided by body height) have significantly decreased in recent years. Even more evident than the changes in pelvic morphology are the rapid changes in body height in most modern countries since the end-19th and particularly since the mid-20th century. Modern Japanese mature earlier; the age at take-off (ATO, the age at which the adolescent growth spurt starts) decreases, and they are taller at all ages. Preece-Baines modelling of six national samples of Japanese children and adolescents, surveyed between 1955 and 2000, shows that this gain in height is largely an adolescent trend, whereas height at take-off (HTO) increased by less than 3 cm since 1955; adolescent growth (height gain between ATO and adult age) increased by 6 cm. The effect of globalization on the modern post-war Japanese society (“community effect in height”) on adolescent growth is discussed. KW - Pelvic breadth KW - Elbow breadth KW - Growth KW - Adult height KW - Community effect in height ER - TY - JOUR TI - Automated Bayesian variable selection methods for binary regression models with missing covariate data AU - Bergrab, Michael AU - Aßmann, Christian PY - 2025 SP - 203-244 PB - Otto-Friedrich-Universität CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/107070 LA - eng AB - Data collection and the availability of large data sets has increased over the last decades. In both statistical and machine learning frameworks, two methodological issues typically arise when performing regression analysis on large data sets. First, variable selection is crucial in regression modeling, as it helps to identify an appropriate model with respect to the considered set of conditioning variables. Second, especially in the context of survey data, handling of missing values is important for estimation, which occur even with state-of-the-art data collection and processing methods. Within this paper, we provide an Bayesian approach based on a spike-and-slab prior for the regression coefficients, which allows for simultaneous handling of variable selection and estimation in combination with handling of missing values in covariate data. The paper also discusses the implementation of the approach using Markov chain Monte Carlo techniques and provides results for simulated data sets and an empirical illustration based on data from the German National Educational Panel Study. The suggested Bayesian approach is compared to other statistical and machine learning frameworks such as Lasso, ridge regression, and Elastic net, and is shown to perform well in terms of estimation performance and variable selection accuracy. The simulation results demonstrate that ignoring the handling of missing values in data sets can lead to the generation of biased selection results. Overall, the proposed Bayesian method offers a holistic, flexible, and powerful framework for variable selection in the presence of missing covariate data. KW - Bayesian Estimation KW - Variable Selection KW - Model Selection KW - Sparsity KW - Spike-and-Slab Prior KW - Elastic Net KW - Shrinkage KW - Estimation ER - TY - JOUR TI - Automated Bayesian variable selection methods for binary regression models with missing covariate data AU - Bergrab, Michael AU - Aßmann, Christian PY - 2024 JO - Wirtschafts- und sozialstatistisches Archiv : ASTA ; eine Zeitschrift der Deutschen Statistischen Gesellschaft VL - 18 IS - 2 SP - 203-244 PB - Springer CY - Berlin ; Heidelberg SN - 1863-8155 DO - 10.1007/s11943-024-00345-1 UR - https://fis.uni-bamberg.de/handle/uniba/107069 LA - eng AB - Data collection and the availability of large data sets has increased over the last decades. In both statistical and machine learning frameworks, two methodological issues typically arise when performing regression analysis on large data sets. First, variable selection is crucial in regression modeling, as it helps to identify an appropriate model with respect to the considered set of conditioning variables. Second, especially in the context of survey data, handling of missing values is important for estimation, which occur even with state-of-the-art data collection and processing methods. Within this paper, we provide an Bayesian approach based on a spike-and-slab prior for the regression coefficients, which allows for simultaneous handling of variable selection and estimation in combination with handling of missing values in covariate data. The paper also discusses the implementation of the approach using Markov chain Monte Carlo techniques and provides results for simulated data sets and an empirical illustration based on data from the German National Educational Panel Study. The suggested Bayesian approach is compared to other statistical and machine learning frameworks such as Lasso, ridge regression, and Elastic net, and is shown to perform well in terms of estimation performance and variable selection accuracy. The simulation results demonstrate that ignoring the handling of missing values in data sets can lead to the generation of biased selection results. Overall, the proposed Bayesian method offers a holistic, flexible, and powerful framework for variable selection in the presence of missing covariate data. KW - Bayesian Estimation KW - Variable Selection KW - Model Selection KW - Sparsity KW - Spike-and-Slab Prior KW - Elastic Net KW - Shrinkage KW - Estimation ER - TY - JOUR TI - Post-processing for Bayesian analysis of reduced rank regression models with orthonormality restrictions AU - Aßmann, Christian AU - Boysen-Hogrefe, Jens AU - Pape, Markus PY - 2024 SP - 577-609 PB - Otto-Friedrich-Universität CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/105513 LA - eng AB - Orthonormality constraints are common in reduced rank models. They imply that matrix-variate parameters are given as orthonormal column vectors. However, these orthonormality restrictions do not provide identification for all parameters. For this setup, we show how the remaining identification issue can be handled in a Bayesian analysis via post-processing the sampling output according to an appropriately specified loss function. This extends the possibilities for Bayesian inference in reduced rank regression models with a part of the parameter space restricted to the Stiefel manifold. Besides inference, we also discuss model selection in terms of posterior predictive assessment. We illustrate the proposed approach with a simulation study and an empirical application. KW - Bayesian estimation KW - Post-processing KW - Reduced rank regression KW - Orthogonal transformation KW - Model selection KW - Stiefel manifold KW - Posterior predictive assessment 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 - Post-processing for Bayesian analysis of reduced rank regression models with orthonormality restrictions AU - Aßmann, Christian AU - Boysen-Hogrefe, Jens AU - Pape, Markus PY - 2024 JO - Advances in statistical analysis : AStA VL - 108 IS - 3 SP - 577-609 PB - Springer CY - Berlin ; Heidelberg SN - 1863-8171 DO - 10.1007/s10182-023-00489-5 UR - https://fis.uni-bamberg.de/handle/uniba/92532 LA - eng AB - Orthonormality constraints are common in reduced rank models. They imply that matrix-variate parameters are given as orthonormal column vectors. However, these orthonormality restrictions do not provide identification for all parameters. For this setup, we show how the remaining identification issue can be handled in a Bayesian analysis via post-processing the sampling output according to an appropriately specified loss function. This extends the possibilities for Bayesian inference in reduced rank regression models with a part of the parameter space restricted to the Stiefel manifold. Besides inference, we also discuss model selection in terms of posterior predictive assessment. We illustrate the proposed approach with a simulation study and an empirical application. KW - Bayesian estimation KW - Post-processing KW - Reduced rank regression KW - Orthogonal transformation KW - Model selection KW - Stiefel manifold KW - Posterior predictive assessment ER - TY - JOUR TI - Chain Reversion for Detecting Associations in Interacting Variables - St. Nicolas House Analysis AU - Aßmann, Christian AU - Hermanussen, Michael AU - Groth, Detlef PY - 2023 SP - 1-14 PB - Otto-Friedrich-Universität CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/58651 LA - eng AB - (1) Background: We present a new statistical approach labeled as “St. Nicolas House Analysis” (SNHA) for detecting and visualizing extensive interactions among variables. (2) Method: We rank absolute bivariate correlation coefficients in descending order according to magnitude and create hierarchic “association chains” defined by sequences where reversing start and end point does not alter the ordering of elements. Association chains are used to characterize dependence structures of interacting variables by a graph. (3) Results: SNHA depicts association chains in highly, but also in weakly correlated data, and is robust towards spurious accidental associations. Overlapping association chains can be visualized as network graphs. Between independent variables significantly fewer associations are detected compared to standard correlation or linear model-based approaches. (4) Conclusion: We propose reversible association chains as a principle to detect dependencies among variables. The proposed method can be conceptualized as a non-parametric statistical method. It is especially suited for secondary data analysis as only aggregate information such as correlations matrices are required. The analysis provides an initial approach for clarifying potential associations that may be subject to subsequent hypothesis testing. KW - St. Nicolas House Analysis KW - association chains KW - bivariate correlation coefficients KW - network graphs KW - data matrices 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 - 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 - Chain Reversion for Detecting Associations in Interacting Variables—St. Nicolas House Analysis AU - Aßmann, Christian AU - Hermanussen, Michael AU - Groth, Detlef PY - 2021 JO - International Journal of Environmental Research and Public Health VL - 18 IS - 4 SP - 1-14 PB - MDPI AG CY - Basel SN - 1660-4601 DO - 10.3390/ijerph18041741 UR - https://fis.uni-bamberg.de/handle/uniba/49508 LA - eng AB - (1) Background: We present a new statistical approach labeled as “St. Nicolas House Analysis” (SNHA) for detecting and visualizing extensive interactions among variables. (2) Method: We rank absolute bivariate correlation coefficients in descending order according to magnitude and create hierarchic “association chains” defined by sequences where reversing start and end point does not alter the ordering of elements. Association chains are used to characterize dependence structures of interacting variables by a graph. (3) Results: SNHA depicts association chains in highly, but also in weakly correlated data, and is robust towards spurious accidental associations. Overlapping association chains can be visualized as network graphs. Between independent variables significantly fewer associations are detected compared to standard correlation or linear model-based approaches. (4) Conclusion: We propose reversible association chains as a principle to detect dependencies among variables. The proposed method can be conceptualized as a non-parametric statistical method. It is especially suited for secondary data analysis as only aggregate information such as correlations matrices are required. The analysis provides an initial approach for clarifying potential associations that may be subject to subsequent hypothesis testing. KW - St. Nicolas House Analysis KW - association chains KW - bivariate correlation coefficients KW - network graphs KW - data matrices 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 influence of regional school infrastructure and labor market conditions on the transition process to secondary schooling in Germany AU - Sixt, Michaela AU - Aßmann, Christian PY - 2020 JO - Journal for educational research online : JERO = Journal für Bildungsforschung online VL - 12 IS - 2 SP - 36-66 PB - Waxmann CY - Münster SN - 1866-6671 UR - https://fis.uni-bamberg.de/handle/uniba/48750 LA - eng ER -