Modeling Dynamics in Space and Time
Data collected over both units (e.g. municipalities, states, countries) and time (e.g. days, months, years)---known as pooled time series---are common in social science. By gaining leverage both across units and over time, this data structure helps us answer important questions that would be difficult if we only looked at a single year (i.e. cross section) or single country (i.e. time series): the relationship between growth and democracy, whether or not the resource curse exists, and how institutions shape political and economic outcomes. However, pooled time series often show types of heterogeneity that make standard regression approaches inappropriate. In this course we will explore why we might want to use pooled time series data, what issues might arise when using data in time and space, and how we can test for violations of assumptions and account for them.
This five-day course is an applied intermediate-level course focusing on the techniques for testing theories with pooled time series data. During the first three days, the course will involve about three hours of lecture time with breaks, then lunch, then three to four hours of hands on instruction in analysis that takes place in smaller groups using STATA. On the fourth day, students will work on a specific project assignment that applies the concepts introduced in the course. On the final day, we will have a wrap-up lecture and discuss group presentations.
This course runs January 22-26,2018.
Day 1: On the first day, we will cover matrix algebra. Matrix algebra is often necessary to understand cross-sectional time series data, since individual countries are “stacked” on one another. No previous experience with matrix algebra is necessary. We will talk about operations with matrices, OLS with matrix algebra, and GLS with matrix algebra.
Day 2: On the second day, we focus on introducing pooled time series data. We will discuss the advantages and disadvantages of pooling multiple units over time. We will also discuss simple models to test for and interpret the pooling assumption.
Day 3: On the third day we will delve into models of pooled time series data used by applied researchers. These include random and fixed effects models, and panel corrected standard errors. We will also expand some of the time series diagnostics, such as unit root testing, to the pooled time series context.
Day 4: The fourth day is a gentle introduction to spatial statistics. Recent models allow us to test for and model heterogeneity across space. We will discuss various approaches to thinking about, estimating, and interpreting spatial models.
Day 5: The last day, we will have student presentations of your research project you have developed over the week. Everyone will provide feedback. If needed, we will also finish up any lectures.
A full-semester graduate-level course in multiple regression analysis, Modeling Dynamics (offered in the IPSA-USP 2017 Summer School) and Advances in Modeling Dynamics (offered in the IPSA-USP 2017 Summer School) or the equivalent background in time series analysis.