Longitudinal and Event History Models for the Social and Political Sciences

3 units

This course will cover the statistical concepts and techniques that are used to model social and political events over time, including basic time-series and event history (survival) data. Such data routinely occurs in both the social sciences and public health sciences. Lectures will introduce: second order stationary time series, autoregressive structures, spectrum and linear filtering theory, autocorrelation consistent (HAC) variance estimation, survival functions, hazard rates, types of censoring and truncation. Modes of inference for regression models will be provided. All applied work will be in the R software environment for statistical computing and graphics. Students will be able to identify and classify data problems in longitudinal analysis, define the appropriate function accounting for time as well as summarize and interpret analyses of such data using various estimators. In addition, participants will able to formulate research questions related to longitudinal data and the appropriate associated regression models or other approach.