Comparing Bayesian Variable Selection to Lasso Approaches for Applications in Psychology

被引:2
|
作者
Bainter, Sierra A. [1 ,4 ]
McCauley, Thomas G. [2 ]
Fahmy, Mahmoud M. [1 ]
Goodman, Zachary T. [1 ]
Kupis, Lauren B. [3 ]
Rao, J. Sunil [1 ]
机构
[1] Univ Miami, Miami, FL USA
[2] Univ Calif San Diego, San Diego, CA 92110 USA
[3] Univ Calif Los Angeles, Los Angeles, CA USA
[4] Univ Miami, Dept Psychol, 5665 Poncede Leon Blvd, Coral Gables, FL 33146 USA
关键词
Bayesian; regression; lasso; variable selection; penalization; shrinkage priors; stochastic search variable selection; MAJOR DEPRESSIVE DISORDER; REGRESSION SHRINKAGE; STEPWISE REGRESSION; MODEL; REGULARIZATION; QUESTIONNAIRE; PERSONALITY; IMPULSIVITY; PREDICTION; ANXIETY;
D O I
10.1007/s11336-023-09914-9
中图分类号
O1 [数学];
学科分类号
0701 ; 070101 ;
摘要
In the current paper, we review existing tools for solving variable selection problems in psychology. Modern regularization methods such as lasso regression have recently been introduced in the field and are incorporated into popular methodologies, such as network analysis. However, several recognized limitations of lasso regularization may limit its suitability for psychological research. In this paper, we compare the properties of lasso approaches used for variable selection to Bayesian variable selection approaches. In particular we highlight advantages of stochastic search variable selection (SSVS), that make it well suited for variable selection applications in psychology. We demonstrate these advantages and contrast SSVS with lasso type penalization in an application to predict depression symptoms in a large sample and an accompanying simulation study. We investigate the effects of sample size, effect size, and patterns of correlation among predictors on rates of correct and false inclusion and bias in the estimates. SSVS as investigated here is reasonably computationally efficient and powerful to detect moderate effects in small sample sizes (or small effects in moderate sample sizes), while protecting against false inclusion and without over-penalizing true effects. We recommend SSVS as a flexible framework that is well-suited for the field, discuss limitations, and suggest directions for future development.
引用
收藏
页码:1032 / 1055
页数:24
相关论文
共 50 条
  • [1] Comparing Bayesian Variable Selection to Lasso Approaches for Applications in Psychology
    Sierra A. Bainter
    Thomas G. McCauley
    Mahmoud M. Fahmy
    Zachary T. Goodman
    Lauren B. Kupis
    J. Sunil Rao
    [J]. Psychometrika, 2023, 88 : 1032 - 1055
  • [2] Dirichlet Lasso: A Bayesian approach to variable selection
    Das, Kiranmoy
    Sobel, Marc
    [J]. STATISTICAL MODELLING, 2015, 15 (03) : 215 - 232
  • [3] Bayesian Variable Selection and Estimation for Group Lasso
    Xu, Xiaofan
    Ghosh, Malay
    [J]. BAYESIAN ANALYSIS, 2015, 10 (04): : 909 - 936
  • [4] Approaches for Bayesian variable selection
    George, EI
    McCulloch, RE
    [J]. STATISTICA SINICA, 1997, 7 (02) : 339 - 373
  • [5] On Bayesian lasso variable selection and the specification of the shrinkage parameter
    Anastasia Lykou
    Ioannis Ntzoufras
    [J]. Statistics and Computing, 2013, 23 : 361 - 390
  • [6] Model uncertainty and variable selection in Bayesian lasso regression
    Chris Hans
    [J]. Statistics and Computing, 2010, 20 : 221 - 229
  • [7] Model uncertainty and variable selection in Bayesian lasso regression
    Hans, Chris
    [J]. STATISTICS AND COMPUTING, 2010, 20 (02) : 221 - 229
  • [8] On Bayesian lasso variable selection and the specification of the shrinkage parameter
    Lykou, Anastasia
    Ntzoufras, Ioannis
    [J]. STATISTICS AND COMPUTING, 2013, 23 (03) : 361 - 390
  • [9] Variable Selection Using the Lasso-Cox Model with Bayesian Regularization
    Lu, Wenxin
    Yu, Zhuliang
    Gu, Zhenghui
    Huang, Jinhong
    Gao, Wei
    Zhou, Haiyu
    [J]. PROCEEDINGS OF THE 2018 13TH IEEE CONFERENCE ON INDUSTRIAL ELECTRONICS AND APPLICATIONS (ICIEA 2018), 2018, : 924 - 927
  • [10] Comparing Spike and Slab Priors for Bayesian Variable Selection
    Malsiner-Walli, Gertraud
    Wagner, Helga
    [J]. AUSTRIAN JOURNAL OF STATISTICS, 2011, 40 (04) : 241 - 264