Extends the linear model to account for temporal dynamics and cross-sectional variation. Focuses on model selection and real-world interpretation of model results. Topics include autoregressive processes, trends, seasonality, stationarity, lagged dependent variables, ARIMA models, fixed effects, random effects, cointegration and error correction models, panel heteroskedasticity, missing data in panel models, causal inference with panel data. Recommended: Graduate level coursework in linear regression and social science research design. Basic familiarity with or willingness to learn the R statistical language. Offered: jointly with POL S 512.
Course Name
Time Series and Panel Data for the Social Sciences
Credits
5
Quarter(s)
Spring