Conducting Multivariate Analyses of Social, Economic, and Political Data (Montreal, QC)


This workshop teaches participants how to conduct multivariate analyses of social, economic, and political data. Several valuable tools for academic researchers in the social sciences and analysts in the public and private sectors are considered. The course first focuses on discrete choice models including binomial and multinomial logit and multinomial probit models and the mixed logit model for analyzing factors affecting choices individuals make in social, economic, and political settings. The course then considers time series methods for analyzing the effects of policy interventions and socio-economic factors on economic, political, and social outcomes. Topics include ARIMA models for interventions and forecasting, error correction models of fully and fractionally cointegrated processes, GARCH models, VAR/VECM models, and Markov switching models. Assuming only familiarity with the standard OLS regression model, the course emphasizes practical approaches to analyzing individual- and aggregate-level data. Participants learn how to use major software packages such as NLOGIT, Stata, R, Rats, and EVIEWS in their research. Participants are invited to bring their own data sets to lab sessions.

Fee: Members = $1500; Non-members = $3000

Tags: multivariate, choice models, dynamic models

Course Sections

Report Problem