Stata Day at Georgia State
This event is only available for GSU Faculty, Staff, and Current Students. Please use your GSU email when registering for the event.
Multilevel/Mixed-effects Models Using Stata
This talk will present an introduction to multilevel (mixed) models and give a demonstration of fitting them in Stata. How to fit both random-intercept and random-coefficient models as well as crossed models and models with special residual structures will be discussed. While this session will be primarily for linear (Gaussian) models, I will briefly consider generalized linear models to show that they are syntactically identical.
Fitting and Interpreting Regression Models with Interactions/Moderation
Stata makes it easy to fit linear regression models with interaction effects and provides convenient postestimation tools to simplify interpretation of these models. In this talk, I will cover the use of Stata's factor variable syntax for including interactions in a model. The main focus of the talk will be the the use of postestimation commands to generate, graph, and test differences in marginal means, marginal effects, and predicted values.
Structural Equation Models (SEM) in Stata
Structural equation modeling encompasses a wide variety of models that can include confirmatory factor analysis, regression models, and models with both latent variables and structural components. This talk provides an overview of Stata's -sem- and -gsem- commands, which can be used to fit SEMs that include a variety of outcome types (continuous, categorical, count). Examples of multilevel and multigroup SEMs will be shown. Tools for evaluating model fit will also be discussed.
Location: 58 Edgewood Ave SE, Room 365
- Thursday, September 6, 2018
- 9:30am - 4:30pm