Analyzing Multilevel and Mixed Models Using Stata

Instructor(s):

• Roberto Gutierrez, Statistics, StataCorp LP

This five day workshop is an introduction to using Stata to fit multilevel mixed models. Mixed models contain both fixed effects analogous to the coefficients in standard regression models and random effects not directly estimated but instead summarized through the unique elements of their variance covariance matrix. Mixed models may contain more than one level of nested random effects and hence these models are also referred to as multilevel or hierarchical models particularly in the social sciences. The first day will cover the history and theory behind the linear mixed model whose origins can be traced to analysis of variance and the estimation of variance components therein. Stata's approach to linear mixed models is to assign random effects to independent panels where a hierarchy of nested panels can be defined for handling nested random effects. Much of the first day will be spent tying this approach to that from classical linear models and becoming familiar with the terminology for both approaches. The second day's focus will be on introducing random coefficients and the various covariance structures that can be imposed with multiple random effects terms. The theme of the third day can best be described as tricks of the trade, covering various methods for fitting more complex models including crossed effects models, growth curve models, and models with complex and grouped constraints on covariance structures. The fourth day will consist of predictions model diagnostics and other post estimation tasks. During the first four days, the discussion will be confined to linear mixed models for continuous responses. The fifth day's focus will be on models for other types of responses in particular binary and count responses. It will be shown that most of what is discussed for linear mixed models can be applied equally to mixed models with non continuous responses. Each day will consist of a morning session consisting of theory and classroom discussion followed by an afternoon session that puts these concepts into practice by performing analyses on real data interactively using Stata.

Dates:  August 17-21 

Fee:  Member: $1800; Non-member: $4000

This course is limited to 20 participants.