Document Type : Original Paper
Authors
Department of Statistics, Tarbiat Modares University, Tehran, Islamic Republic of Iran
Abstract
When discussing non-Gaussian spatially correlated variables, generalized linear mixed models have enough flexibility for modeling various data types. However, the maximum likelihood methods are plagued with substantial calculations for large data sets, resulting in long waiting times for estimating the model parameters. To alleviate this drawback, composite likelihood functions obtained from the product of the likelihoods of subsets of observations are used. The current paper uses the pairwise likelihood method to study the parameter estimations of spatial generalized linear mixed models. Then, we use the weighted pairwise and penalized likelihood functions to estimate the parameters of the mentioned models. The accuracy of estimates based on these likelihood functions is evaluated and compared with full likelihood function-based estimation using simulation studies. Based on our results, the penalized likelihood function improved parameter estimation. Prediction using penalized likelihood functions is applied. Ultimately, pairwise and penalized pairwise likelihood methods are applied to analyze count real data sets.
Keywords
- Spatial generalized linear model
- Composite likelihood
- Penalized pairwise likelihood
- Weighted pairwise likelihood
Main Subjects
- Nelder, JA and Wedderburn, RWM. Generalized Linear Models: Journal of the Royal Statistical Association, Series A. 1972;135: 370-384.
- McCullagh, CE and Nelder, JA. Generalized Linear Models: Chapman and Hall, London 1989.
- McCulloch, C. Maximum Likelihood Algorithms for Generalized Linear Mixed Models: Journal of the American Statistical Association. 1997; 92: 162-170.
- Mohammadzadeh, N. and Hosseini, F. Maximum-Likelihood Estimation for Spatial GLMMs: Procedia Environmental Sciences. 2011; 3: 63-68.
- Baghishani, H and Mohammadzadeh, M. A Data Cloning Algorithm for Computing Maximum Likelihood Estimates in Spatial Generalized Linear Mixed Models: Computational Statistics and Data Analysis. 2011; 55: 1748-1759.
- Torabi, M. Likelihood Inference for Spatial Generalized Linear Mixed Models: Communications in Statistics-Simulation and Computation. 2015; 44: 1692-1701.
- Rue, H and Martino, S. Approximate Bayesian Inference for Hierarchical Gaussian Markov Random Field Models: Journal of Statistical Planning and Inference. 2007; 137: 3177-3192.
- Eidsvik, J, Finley, AO, Banerjee, S and Rue, H. Approximate Bayesian Inference for Large Spatial Datasets Using Predictive Process Models: Computational Statistics and Data Analysis. 2012; 56: 1362-1380.
- McCullagh, P. Quasi-Likelihood Functions: Annals of Statistics. 1983; 11: 59-67.
- Wedderburn, R. Quasi-Likelihood Functions, Generalized Linear Models and the Gauss-Newton Method: Biometrica. 1974; 61: 973-981.
- Heagerty, P. and Lele, S. A Composite Likelihood Approach to Binary Spatial Data: Journal of American Statistical Association. 1998; 93: 1099-1111.
- Bevilacqua, M and Gaetan, C. Comparing composite likelihood methods based on pairs for spatial Gaussian random fields: Statistics and Computing. 2015; 25: 877-892.
- Mazo, G, Karlis, D and Rau, A. A randomized pairwise likelihood method for complex statistical inferences: Journal of the American Statistical Association. 2024; 119(547): 2317-2327.
- Cox, D and Reid, NA. Note on Pseudo Likelihood Constructed from Marginal Densities: Biometrika. 2004; 91: 729-737.
- Varin, C, Host, G. and Skare, O. Pairwise Likelihood Inference in Spatial Generalized Linear Mixed Models: Computational Statistics and Data Analysis. 2005; 49: 1173-1191.
- Bevilacqua, M, Mateu, J, Porcu, E, Zhang, H and Zini, A. Weighted Composite Likelihood-Based Tests for Space-Time Separability of Covariance Functions: Statistics and Computing. 2010; 20: 283-293.
- Joe, H and Lee, Y. On Weighting of Bivariate Margins in Pairwise Likelihood: Journal of Multivariate Analysis. 2009; 100: 670-685.
- Li, Q, Sun, X, Wang, N and Gao, X. Penalized composite likelihood for colored graphical Gaussian models: Statistical Analysis and Data Mining: The ASA Data Science Journal. 2021;14(4): 366-378.
- Bonat, WH and Ribeiro, PJ. Practical Likelihood Analysis for Spatial Generalized Linear Mixed Models: Environmetrics. 2016; 27: 83-89.
- Lindsay, B. Composite Likelihood Methods: Contemporary Mathematics. 1988; 80: 220-239.
- Azzalini, A and Arellano-Valle, RB. Maximum Penalized Likelihood Estimation for Skew-Normal and Skew-t Distributions: Journal of Statistical Planning and Inference. 2013; 143:419-433.
- Smith, B, Wang, S, Wong, A and Zhou, X. A penalized likelihood approach to parameter estimation with integral reliability constraints: Entropy. 2015; 17(6): 4040-4063.
- Firth, D. Bias Reduction of Maximum Likelihood Estimates: Biometrika. 1993; 80: 27-38.
- Tibshirani, R. Regression Shrinkage and Selection via the Lasso: Journal of the Royal Statistical Society, Series B(Methodological). 1996; 267-288.
|
|
- Green, PJ. On Use of the EM for Penalized Likelihood Estimation: Journal of the Royal Statistical Society, Series B (Methodological). 1990; 443-452.
- Booth, JG and Hobert, JP. Maximizing Generalized Linear Mixed Model Likelihoods with an Automated Monte Carlo EM Algorithm: Royal Statistical Society, Series B. 1999; 61: 265-285.
- Zhang, H. On Estimation and Prediction for Spatial Generalized Linear Mixed Models, Biometrics. 2002; 58: 129-136.
- Guillot, G, Loren, N, Rudemo, M. Spatial Prediction of Weed Intensities from Exact Count Data and Image-Based Estimates, Journal of the Royal Statistical Society. 2009; 58: 525-542.
)