Multilevel/Mixed Models Using Stata

Description

This two-day course 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. 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.

The course will be taught in five parts. The first part of the course will look at the classic random-intercept linear model. We will discuss several approaches for fitting this model, along with the associated benefits and assumptions of each approach. The second part will focus on random coefficients and the various covariance structures that can be imposed with multiple random-effects terms. The theme of the third part 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 part will consist of predictions, model diagnostics, and other postestimation tasks. During the first four parts, the discussion will be confined to linear mixed models for continuous responses. The fifth part will focus on models for other types of responses in particular binary and count responses. During this part of the course, you will learn that most of what is discussed for linear mixed models can be applied equally to mixed models with noncontinuous responses.

The course will be interactive, use real data, and offer ample opportunity for specific research questions and for working exercises to enforce what is learned. Enroll now.

Course topics

• Part I
  • What constitutes a linear mixed model?
  • The random-intercept model
  • The within estimator versus the GLS estimator; the Hausman test
  • Maximum likelihood and restricted maximum likelihood
  • Using xtmixed and xtreg for the random-intercept model
• Part II
  • Adding random coefficients
  • Specifying models hierarchically
  • Covariance structures for random effects
  • Growth curves
  • Linear transformations of covariates in a random-effects setting
  • Likelihood ratio (LR) tests
• Part III
  • Multiple-level models
  • Crossed-effects models
  • Using Stata’s “R.” factor notation for mixed models
  • Complex and grouped constraints on variance components
  • Heteroskedastic residual errors
  • Alternate residual-error structures
• Part IV
  • Best linear unbiased predictions (BLUPs)
  • Residuals
  • Fit diagnostics
  • Diagnostic plots
  • Cataloging and comparing mixed-models results in Stata
• Part V
  • Binary and count responses
  • Estimation via adaptive Gaussian quadrature
  • Model building using the Laplacian approximation
  • Predictions and other postestimation tasks

Prerequisite

• Basic knowledge of standard linear regression and a working knowledge of Stata and the Do-File Editor

Next session

The next available session will be held September 16–17, in San Francisco at MicroTek. Other locations and dates can be found here. The price is $1,295 per person. Enroll now.

Click here to request to be notified of future training sessions.

Notes

Enrollment is limited to 24 participants. Computers with Stata installed are provided at all public training sessions. All training courses run from 8:30 AM to 4:30 PM each day. A continental breakfast, lunch, and an afternoon snack will also be provided; the breakfast is available before the course begins. All participants are encouraged to bring a USB flash drive to all public training sessions; this is the safest and simplest way to save your work from the session.