Generalized linear mixed models (GLMMs) are typically constructed by incorporating random effects into the linear predictor. The random effects are usually assumed to be normally distributed with mean zero and varianc...Generalized linear mixed models (GLMMs) are typically constructed by incorporating random effects into the linear predictor. The random effects are usually assumed to be normally distributed with mean zero and variance-covariance identity matrix. In this paper, we propose to release random effects to non-normal distributions and discuss how to model the mean and covariance structures in GLMMs simultaneously. Parameter estimation is solved by using Quasi-Monte Carlo (QMC) method through iterative Newton-Raphson (NR) algorithm very well in terms of accuracy and stabilization, which is demonstrated by real binary salamander mating data analysis and simulation studies.展开更多
Joint location and scale models of the skew-normal distribution provide useful ex- tension for joint mean and variance models of the normal distribution when the data set under consideration involves asymmetric outcom...Joint location and scale models of the skew-normal distribution provide useful ex- tension for joint mean and variance models of the normal distribution when the data set under consideration involves asymmetric outcomes. This paper focuses on the maximum likelihood estimation of joint location and scale models of the skew-normal distribution. The proposed procedure can simultaneously estimate parameters in the location model and the scale model. Simulation studies and a real example are used to illustrate the proposed methodologies.展开更多
The Bertalanffy-Pütter (BP) five-parameter growth model provides a versatile framework for the modeling of growth. Using data from a growth experiment in literature about the average size-at-age of 24 species of ...The Bertalanffy-Pütter (BP) five-parameter growth model provides a versatile framework for the modeling of growth. Using data from a growth experiment in literature about the average size-at-age of 24 species of tropical trees over ten years in the same area, we identified their best-fit BP-model parameters. While different species had different best-fit exponent-pairs, there was a model with a good fit to 21 (87.5%) of the data </span><span style="font-family:Verdana;">(</span><span style="font-family:""><span style="font-family:Verdana;">“Good fit” means a </span><span style="font-family:Verdana;">normalized root-mean-squared-error <i></span><i><span style="font-family:Verdana;">NRMSE</span></i><span style="font-family:Verdana;"></i> below 2.5%. This threshold was the 95% quantile of the lognormal distribution that was fitted to the <i></span><i><span style="font-family:Verdana;">NRMSE</span></i><span style="font-family:Verdana;"></i> values for the best-fit models for the data)</span></span><span style="font-family:Verdana;">.</span><span style="font-family:Verdana;"> In view of the sigmoidal character of this model despite the early stand we discuss </span><span style="font-family:Verdana;">whether </span><span style="font-family:Verdana;">the setting of the growth experiment may have impeded growth.展开更多
文摘Generalized linear mixed models (GLMMs) are typically constructed by incorporating random effects into the linear predictor. The random effects are usually assumed to be normally distributed with mean zero and variance-covariance identity matrix. In this paper, we propose to release random effects to non-normal distributions and discuss how to model the mean and covariance structures in GLMMs simultaneously. Parameter estimation is solved by using Quasi-Monte Carlo (QMC) method through iterative Newton-Raphson (NR) algorithm very well in terms of accuracy and stabilization, which is demonstrated by real binary salamander mating data analysis and simulation studies.
基金Supported by the National Natural Science Foundation of China(11261025,11201412)the Natural Science Foundation of Yunnan Province(2011FB016)the Program for Middle-aged Backbone Teacher,Yunnan University
文摘Joint location and scale models of the skew-normal distribution provide useful ex- tension for joint mean and variance models of the normal distribution when the data set under consideration involves asymmetric outcomes. This paper focuses on the maximum likelihood estimation of joint location and scale models of the skew-normal distribution. The proposed procedure can simultaneously estimate parameters in the location model and the scale model. Simulation studies and a real example are used to illustrate the proposed methodologies.
文摘The Bertalanffy-Pütter (BP) five-parameter growth model provides a versatile framework for the modeling of growth. Using data from a growth experiment in literature about the average size-at-age of 24 species of tropical trees over ten years in the same area, we identified their best-fit BP-model parameters. While different species had different best-fit exponent-pairs, there was a model with a good fit to 21 (87.5%) of the data </span><span style="font-family:Verdana;">(</span><span style="font-family:""><span style="font-family:Verdana;">“Good fit” means a </span><span style="font-family:Verdana;">normalized root-mean-squared-error <i></span><i><span style="font-family:Verdana;">NRMSE</span></i><span style="font-family:Verdana;"></i> below 2.5%. This threshold was the 95% quantile of the lognormal distribution that was fitted to the <i></span><i><span style="font-family:Verdana;">NRMSE</span></i><span style="font-family:Verdana;"></i> values for the best-fit models for the data)</span></span><span style="font-family:Verdana;">.</span><span style="font-family:Verdana;"> In view of the sigmoidal character of this model despite the early stand we discuss </span><span style="font-family:Verdana;">whether </span><span style="font-family:Verdana;">the setting of the growth experiment may have impeded growth.
