On a Robust MaxEnt Process Regression Model with Sample-Selection

被引:7
|
作者
Kim, Hea-Jung [1 ]
Bae, Mihyang [1 ]
Jin, Daehwa [1 ]
机构
[1] Dongguk Univ Seoul, Dept Stat, Pil Dong 3Ga, Seoul 100715, South Korea
来源
ENTROPY | 2018年 / 20卷 / 04期
基金
新加坡国家研究基金会;
关键词
Gaussian process model; hierarchical Bayesian methodology; robust sample-selection MaxEnt process regression model; Markov chain Monte Carlo; sample-selection bias; bias collection; SCALE MIXTURES; NORMAL-DISTRIBUTIONS; INFERENCE; CONSTRAINT; ERROR;
D O I
10.3390/e20040262
中图分类号
O4 [物理学];
学科分类号
0702 ;
摘要
In a regression analysis, a sample-selection bias arises when a dependent variable is partially observed as a result of the sample selection. This study introduces a Maximum Entropy (MaxEnt) process regression model that assumes a MaxEnt prior distribution for its nonparametric regression function and finds that the MaxEnt process regression model includes the well-known Gaussian process regression (GPR) model as a special case. Then, this special MaxEnt process regression model, i.e., the GPR model, is generalized to obtain a robust sample-selection Gaussian process regression (RSGPR) model that deals with non-normal data in the sample selection. Various properties of the RSGPR model are established, including the stochastic representation, distributional hierarchy, and magnitude of the sample-selection bias. These properties are used in the paper to develop a hierarchical Bayesian methodology to estimate the model. This involves a simple and computationally feasible Markov chain Monte Carlo algorithm that avoids analytical or numerical derivatives of the log-likelihood function of the model. The performance of the RSGPR model in terms of the sample-selection bias correction, robustness to non-normality, and prediction, is demonstrated through results in simulations that attest to its good finite-sample performance.
引用
收藏
页数:19
相关论文
共 50 条
  • [21] A note on model selection for small sample regression
    Masanori Kawakita
    Jun’ichi Takeuchi
    [J]. Machine Learning, 2017, 106 : 1839 - 1862
  • [22] Outlier robust model selection in linear regression
    Müller, S
    Welsh, AH
    [J]. JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 2005, 100 (472) : 1297 - 1310
  • [23] Algorithms for robust model selection in linear regression
    Morgenthaler, S
    Welsch, RE
    Zenide, A
    [J]. THEORY AND APPLICATION OF RECENT ROBUST METHODS, 2004, : 195 - 206
  • [24] Robust variable selection in the logistic regression model
    Jiang, Yunlu
    Zhang, Jiantao
    Huang, Yingqiang
    Zou, Hang
    Huang, Meilan
    Chen, Fanhong
    [J]. HACETTEPE JOURNAL OF MATHEMATICS AND STATISTICS, 2021, 50 (05): : 1572 - 1582
  • [25] Consistency of Heckman-type two-step estimators for the multivariate sample-selection model
    Tauchmann, Harald
    [J]. APPLIED ECONOMICS, 2010, 42 (30) : 3895 - 3902
  • [26] Process-monitoring-for-quality-A robust model selection criterion for the logistic regression algorithm
    Escobar, Carlos A.
    Morales-Menendez, Ruben
    [J]. MANUFACTURING LETTERS, 2019, 22 : 6 - 10
  • [27] Model Selection for Gaussian Process Regression
    Gorbach, Nico S.
    Bian, Andrew An
    Fischer, Benjamin
    Bauer, Stefan
    Buhmann, Joachim M.
    [J]. PATTERN RECOGNITION (GCPR 2017), 2017, 10496 : 306 - 318
  • [28] An empirical study of sample-selection bias in indices of commercial real estate
    Munneke, HJ
    Slade, BA
    [J]. JOURNAL OF REAL ESTATE FINANCE AND ECONOMICS, 2000, 21 (01): : 45 - 64
  • [29] HOUSEHOLD ALCOHOL AND TOBACCO EXPENDITURES IN TURKEY: A SAMPLE-SELECTION SYSTEM APPROACH
    Bilgic, Abdulbaki
    Yen, Steven T.
    [J]. CONTEMPORARY ECONOMIC POLICY, 2015, 33 (03) : 571 - 585
  • [30] An Empirical Study of Sample-Selection Bias in Indices of Commercial Real Estate
    Henry J. Munneke
    Barrett A. Slade
    [J]. The Journal of Real Estate Finance and Economics, 2000, 21 : 45 - 64