Post-selection inference for linear mixed model parameters using the conditional Akaike information criterion

We investigate the issue of post-selection inference for a fixed and a mixed parameter in a linear mixed model using a conditional Akaike information criterion as a model selection procedure. Within the framework of linear mixed models we develop complete theory to construct confidence intervals for regression and mixed parameters under three frameworks: nested and general model sets as well as misspecified models. Our theoretical analysis is accompanied by a simulation experiment and a post-selection examination on mean income across Galicia’s counties. Our numerical studies confirm a good performance of our new procedure. Moreover, they reveal a startling robustness to the model misspecification of a naive method to construct the confidence intervals for a mixed parameter which is in contrast to our findings for the fixed parameters.

[1]  Nils Lid Hjort,et al.  Focused model selection for linear mixed models with an application to whale ecology , 2020, The Annals of Applied Statistics.

[2]  S. Greven,et al.  On the behaviour of marginal and conditional AIC in linear mixed models , 2010 .

[3]  Hua Liang,et al.  A Note on Conditional AIC for Linear Mixed-Effects Models. , 2008, Biometrika.

[4]  Victor Chernozhukov,et al.  Uniform post-selection inference for least absolute deviation regression and other Z-estimation problems , 2013, 1304.0282.

[5]  M. Woodroofe.  On Model Selection and the ARC Sine Laws , 1982 .

[6]  John T. Ormerod,et al.  On generalized degrees of freedom with application in linear mixed models selection , 2016, Stat. Comput..

[7]  H. Akaike,et al.  Information Theory and an Extension of the Maximum Likelihood Principle , 1973 .

[8]  Tatsuya Kubokawa,et al.  Modified conditional AIC in linear mixed models , 2014, J. Multivar. Anal..

[9]  D. Harville.  Maximum Likelihood Approaches to Variance Component Estimation and to Related Problems , 1977 .

[10]  F. Vaida,et al.  Conditional Akaike information for mixed-effects models , 2005 .

[11]  D. Morales,et al.  Poverty mapping in small areas under a twofold nested error regression model , 2017 .

[12]  N. Hjort,et al.  Frequentist Model Average Estimators , 2003 .

[13]  H. Leeb,et al.  CAN ONE ESTIMATE THE UNCONDITIONAL DISTRIBUTION OF POST-MODEL-SELECTION ESTIMATORS? , 2003, Econometric Theory.

[14]  Maximilian Kasy,et al.  Uniformity and the Delta Method , 2015, Journal of Econometric Methods.

[15]  Domingo Morales,et al.  A Course on Small Area Estimation and Mixed Models , 2021 .

[16]  B. Francq,et al.  Confidence, prediction, and tolerance in linear mixed models , 2019, Statistics in medicine.

[17]  Shonosuke Sugasawa,et al.  Observed best selective prediction in small area estimation , 2019, J. Multivar. Anal..

[18]  A. Buja,et al.  Valid post-selection inference , 2013, 1306.1059.

[19]  Tatsuya Kubokawa,et al.  Conditional and unconditional methods for selecting variables in linear mixed models , 2011, J. Multivar. Anal..

[20]  J. Hodges,et al.  Counting degrees of freedom in hierarchical and other richly-parameterised models , 2001 .

[21]  Halbert White,et al.  Estimation, inference, and specification analysis , 1996 .

[22]  Tatsuya Kubokawa,et al.  Conditional information criteria for selecting variables in linear mixed models , 2008, J. Multivar. Anal..

[23]  F. Gumedze,et al.  Parameter estimation and inference in the linear mixed model , 2011 .

[24]  María José Lombardía,et al.  Empirical best prediction under area-level Poisson mixed models , 2016 .

[25]  C. R. Henderson,et al.  Best linear unbiased estimation and prediction under a selection model. , 1975, Biometrics.

[26]  S. Sperlich,et al.  Simultaneous inference for linear mixed model parameters with an application to small area estimation , 2019, International Statistical Review.

[27]  J. Ware,et al.  Random-effects models for longitudinal data. , 1982, Biometrics.

[28]  M.,et al.  THE FINITE-SAMPLE DISTRIBUTION OF POST-MODEL-SELECTION ESTIMATORS AND UNIFORM VERSUS NONUNIFORM APPROXIMATIONS , 2003, Econometric Theory.

[29]  María José Lombardía,et al.  Mixed generalized Akaike information criterion for small area models , 2017 .

[30]  Donald E. Myers,et al.  Linear and Generalized Linear Mixed Models and Their Applications , 2008, Technometrics.

[31]  Q. Vuong.  Likelihood Ratio Tests for Model Selection and Non-Nested Hypotheses , 1989 .

[32]  S. Müller,et al.  Model Selection in Linear Mixed Models , 2013, 1306.2427.

[33]  B. M. Pötscher,et al.  CAN ONE ESTIMATE THE UNCONDITIONAL DISTRIBUTION OF POST-MODEL-SELECTION ESTIMATORS? , 2007, Econometric Theory.

[34]  N. L. Johnson,et al.  Linear Statistical Inference and Its Applications , 1966 .

[35]  Dennis L. Sun,et al.  Exact post-selection inference, with application to the lasso , 2013, 1311.6238.

[36]  Mollie E. Brooks,et al.  Generalized linear mixed models: a practical guide for ecology and evolution. , 2009, Trends in ecology & evolution.

[37]  H. White,et al.  A Unified Theory of Estimation and Inference for Nonlinear Dynamic Models , 1988 .

[38]  Robin Thompson,et al.  Average information REML: An efficient algorithm for variance parameter estimation in linear mixed models , 1995 .

[39]  Gerda Claeskens,et al.  Asymptotic post‐selection inference for the Akaike information criterion , 2018, Biometrika.

[40]  Ana Ivelisse Avilés,et al.  Linear Mixed Models for Longitudinal Data , 2001, Technometrics.

[41]  F. Bachoc,et al.  Valid confidence intervals for post-model-selection predictors , 2014, The Annals of Statistics.

[42]  Danny Pfeffermann,et al.  Small Area Estimation , 2011, International Encyclopedia of Statistical Science.

[43]  R. Fay,et al.  Estimates of Income for Small Places: An Application of James-Stein Procedures to Census Data , 1979 .

[44]  Yuhong Yang,et al.  Confidence sets for model selection by F -testing , 2015 .

[45]  C. McCulloch,et al.  Misspecifying the Shape of a Random Effects Distribution: Why Getting It Wrong May Not Matter , 2011, 1201.1980.

[46]  R. Tibshirani,et al.  Uniform asymptotic inference and the bootstrap after model selection , 2015, The Annals of Statistics.

[47]  A. Rao,et al.  Estimation of Genetic Parameters: principles , 2003 .