作者: Jay Verkuilen , Michael Smithson
关键词: Regression analysis 、 Applied mathematics 、 Beta distribution 、 Markov chain Monte Carlo 、 Mathematics 、 Statistics 、 Generalized linear model 、 Mixture model 、 Probability distribution 、 Bounded function 、 Normal distribution
摘要: Doubly bounded continuous data are common in the social and behavioral sciences. Examples include judged probabilities, confidence ratings, derived proportions such as percent time on task, scale scores. Dependent variables of this kind often difficult to analyze using normal theory models because their distributions may be quite poorly modeled by distribution. The authors extend beta-distributed generalized linear model (GLM) proposed Smithson Verkuilen (2006) discrete mixtures beta distributions, which enables modeling dependent structures commonly found real settings. discuss estimation both deterministic marginal maximum likelihood stochastic Markov chain Monte Carlo (MCMC) methods. results illustrated three sets from cognitive psychology experiments.