Non-parametric generalized linear mixed models in small area estimation

被引:9
|
作者
Torabi, Mahmoud [1 ]
Shokoohi, Farhad [2 ]
机构
[1] Univ Manitoba, Dept Community Hlth Sci, Winnipeg, MB R3T 2N2, Canada
[2] Ohio State Univ, Dept Stat, Columbus, OH 43210 USA
基金
加拿大自然科学与工程研究理事会;
关键词
Bayesian computation; exponential family; penalized spline; prediction interval; random effect; small area estimation; M-QUANTILE REGRESSION; LIKELIHOOD INFERENCE; DATA CLONING; PREDICTION; ERROR;
D O I
10.1002/cjs.11236
中图分类号
O21 [概率论与数理统计]; C8 [统计学];
学科分类号
020208 ; 070103 ; 0714 ;
摘要
Mixed models are commonly used for the analysis of small area estimation. In particular, small area estimation has been extensively studied under linear mixed models. Recently, small area estimation under the linear mixed model with penalized spline (P-spline) regression model, for fixed part of the model, has been proposed. However, in practice there are many situations that we have counts or proportions in small areas; for example a dataset on the number of asthma physician visits in small areas in Manitoba. In particular, the covariates age, genetic, environmental factors, among other covariates seem to predict asthma physician visits, however, these relationships may not be linear (see Section 5). In this paper, small area estimation under generalized linear mixed models using P-spline regression models is proposed to cover Normal and non-Normal responses. In particular, the empirical best predictor of small area parameters with corresponding prediction intervals are studied. The performance of the proposed approach is evaluated through simulation studies and also by a real dataset. The Canadian Journal of Statistics 43: 82-96; 2015 (c) 2015 Statistical Society of Canada Resume Lorsqu'il est question d'estimation sur des petits domaines, les modeles mixtes sont couramment utilises, et en particulier les modeles lineaires mixtes dont l'usage dans ce contexte a ete largement etudie. Une approche recemment proposee utilise d'ailleurs un modele lineaire mixte dont la partie fixe est un modele de regression base sur des splines penalisees. En pratique, il arrive couramment que les valeurs observees sur les petits domaines soient des variables discretes ou des proportions, notamment dans un jeu de donnees portant sur le nombre de visites chez le medecin pour l'asthme dans differentes regions du Manitoba. Les auteurs constatent que des variables portant sur l'age, la genetique et les facteurs environnementaux semblent de bons predicteurs du nombre de visites chez le medecin, mais que les relations associees a ces variables ne sont pas necessairement lineaires. Les auteurs proposent de proceder a l'estimation pour des petits domaines a l'aide d'un modele lineaire mixte generalise base sur une regression par splines penalisees s'adaptant a une variable reponse normale ou non. Ils etudient les meilleurs estimateurs empiriques pour petits domaines ainsi que les intervalles de prevision leur etant associes. Finalement, ils evaluent la performance de la methode proposee a l'aide de simulations et de l'analyse d'un jeu de donnees reelles. La revue canadienne de statistique 43: 82-96; 2015 (c) 2015 Societe statistique du Canada
引用
收藏
页码:82 / 96
页数:15
相关论文
共 50 条
  • [1] Robust generalized linear mixed models for small area estimation
    Maiti, T
    JOURNAL OF STATISTICAL PLANNING AND INFERENCE, 2001, 98 (1-2) : 225 - 238
  • [2] Robust small area estimation in generalized linear mixed models
    Sanjoy K. Sinha
    METRON, 2019, 77 : 201 - 225
  • [3] Spatial generalized linear mixed models in small area estimation
    Torabi, Mahmoud
    CANADIAN JOURNAL OF STATISTICS-REVUE CANADIENNE DE STATISTIQUE, 2019, 47 (03): : 426 - 437
  • [4] Robust small area estimation in generalized linear mixed models
    Sinha, Sanjoy K.
    METRON-INTERNATIONAL JOURNAL OF STATISTICS, 2019, 77 (03): : 201 - 225
  • [5] Small area estimation under Fay-Herriot models with non-parametric estimation of heteroscedasticity
    Gonzalez-Manteiga, W.
    Lombardia, M. J.
    Molina, I.
    Morales, D.
    Santamaria, L.
    STATISTICAL MODELLING, 2010, 10 (02) : 215 - 239
  • [6] Small area estimation under area-level generalized linear mixed models
    Faltys, Ondrej
    Hobza, Tomas
    Morales, Domingo
    COMMUNICATIONS IN STATISTICS-SIMULATION AND COMPUTATION, 2022, 51 (12) : 7404 - 7426
  • [7] Conditional Non-parametric Bootstrap for Non-linear Mixed Effect Models
    Comets, Emmanuelle
    Rodrigues, Christelle
    Jullien, Vincent
    Ursino, Moreno
    PHARMACEUTICAL RESEARCH, 2021, 38 (06) : 1057 - 1066
  • [8] Conditional Non-parametric Bootstrap for Non-linear Mixed Effect Models
    Emmanuelle Comets
    Christelle Rodrigues
    Vincent Jullien
    Moreno Ursino
    Pharmaceutical Research, 2021, 38 : 1057 - 1066
  • [9] Small Area Estimation Using the Generalized Linear Models
    Hradilova, Jana
    SPSM 2010: STOCHASTIC AND PHYSICAL MONITORING SYSTEMS, 2010, : 101 - 112
  • [10] Generalized linear models for small-area estimation
    Ghosh, M
    Natarajan, K
    Stroud, TWF
    Carlin, BP
    JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1998, 93 (441) : 273 - 282