Panel Data, Fixed Effects, and Instrumental Variables
This topic deals with a variety of different tools, skills and strategies that researchers can use to better understand data and draw conclusions. Students learn about panel data. Panel data is data that examines multiple subjects over multiple time periods. They also learn about instrumental variable research methods, which are sometimes used to estimate causal relationships when random experiments are not possible.
Topic Learning Outcome: Students will have a sophisticated understanding of key analytical techniques including panel data analysis, fixed effect models and regression using instrumental variables. Students will be able to identify instances when each of these techniques is likely to be useful for conducting social science research.
Core Concepts associated with this Topic: Time-Series Model; Bootstrapping; Panel Data; Time Series Data.
University of Toronto: PPG-1004
Stock, James H. and Mark W. Watson. 2011. Introduction to Econometrics, 3rd ed. Pearson/Addison-Wesley. Chapter 10 (p. 347-371) and Chapter 12 (p. 419-456).
Johnson Shoyama Graduate School of Public Policy: JSGS-803
Kahane, Leo H. Regression Basics. Thousand Oaks: Sage, 2008. Chapter 5 (pp. 103-118) and Chapter 6 (pp. 132-138).
François Pétry, Louis M. Imbeau, Jean Crête, and Michel Clavet, “Explaining the Evolution of Government Size in the Canadian Provinces,” Public Finance Review 28:1 (Jan., 2000), pp. 26-47.
Possible Assessment Questions:
Page Created By: Ben Eisen and Joshua Tan, last edited 4 November, 2014.
1.) What is a fixed effect model? When is the fixed effect model useful?
2.) What is panel data? Why is panel data valuable and what types of research questions can it be used to analyze?
3.) What is an instrumental variable? Why do researchers make use of instrumental variables?