Regression Analysis Instructor: Professor K. Bratton
Time: Friday, 9-12
Office Hours: Friday, Wednesday, Thursday 11-12
Office Phone: 578-1912
Home Phone: 343-9820
Email: email@example.com Course Objectives
In this course, we will study multiple regression in depth, focusing on the theoretical foundations and the practical applications of regression analysis. We will begin with an in-depth review of bivariate regression, and move from there to multiple regression. We will examine a number of problems encountered in regression analysis, including multicollinearity, non-linear relationships, non-interval independent variables, heteroscedasticity, and autocorrelation. Toward the end of class, we may introduce advanced topics.
Course grades depend on the following:
1. Four problem sets (to be assigned) (the first worth 10% of the final grade, the following three each worth 15% of the final grade).
Problem sets involve not only generating computer output, but also (and more importantly) interpreting and evaluating results. Students have the option of using any of the computer statistical packages to which they have access. Students are required to turn in both their program syntax and their results.
The first problem set will be distributed in approximately 2 weeks.
2. A midterm (worth 20% of the final grade) and a final (worth 25% of the final grade). These will be closed book, but students will be allowed to bring in one sheet of notes.
Gujarati, Damodar. 1995. Basic Econometrics (3rd ed. Or 4th ed.) New York: McGraw Hill.
Berry, William D. and Stanley Feldman. 1985. Multiple Regression in Practice. Sage Publications #50.
Lewis-Beck, Michael S. 1980. Applied Regression: An Introduction. Sage Publications #22.
Jaccard, James, Robert Turrisi, and Choi K. Wan. Interaction Effects in Multiple Regression. Sage Publications #72.
(assigned reading is to be done in preparation for that class)
Friday, January 21st Introduction
Friday, January 28th Review of 7962. Begin review of bivariate regression
Friday, February 4th Continue review of bivariate regression: Fitting a line, OLS assumptions, estimating a and ß. The estimated slope coefficient "b": variance of b, confidence interval for b, and hypothesis testing of b.
Lewis-Beck: p. 9-20, 26-38, 20-25.
Friday, February 11th Problem Set #1 Assigned.
Residuals as well as explained, unexplained, and total deviations and sums of squares. Standardized variables and standardized variable coefficients; regression forced through the origin. Functional transformations of independent variables. Interpolation, predictive intervals, extrapolation, and outliers. Aggregation bias. Diagnostic plots.
Gujarati: Sections 6.3-6.9, 5.10-5.13
Lewis-Beck: p. 38-47
King, Gary. 1986. "How Not To Lie with Statistics: Avoiding Common Mistakes in Quantitative Political Science." American Journal of Political Science 30: 666-687.
Lewis-Beck, Michael and Andrew Skalaban. 1991. "The R-Squared: Some Straight Talk." In Political Analysis, Vol. 2. Ann Arbor: The University of Michigan Press. Pp. 153-171.
Friedrich, Robert J. 1982. "In Defense of Multiplicative Terms in Multiple Regression Equations." American Journal of Political Science 26: 797-833.
Gill, Jeff. 1999. “Field Essay: The Insignificance of Null Hypothesis Significance Testing.” Political Research Quarterly 52(3).
Friday, March 4th Draft of Problem Set #2 due. Review.
Friday, March 11th
Friday, March 18th Problem Set #2 Due.
Multicollinearity and Multicollinearity diagnostics; dummy and categorical independent variables.
Gujarati, Chapter 10, Sections 15.1-15.5, 15.9
Lewis-Beck, p. 58-63, 66-71
Berry and Feldman, sections 4 and 5
Friday, April 1st Functional Transformations, model specification, missing data, measurement errors.
Gujarati, Sec. 7.7, 7.9-7.12, Chapter 13
Berry and Feldman, sections 2 and 3
Lewis-Beck, p. 56-58, 63-66
Friday, April 8th Regression diagnostics and graphical techniques; analysis of variance and the f-test. Problem set #3 assigned. A closer look at categorical independent variables and f-tests.
Gujarati, Sec. 8.5, 8.6-8.8, 15.6-15.8, 15.11, 15.15
Lewis-Beck, p. 71-74
Bollen, Kenneth and Robert Jackman. 1990. "Regression Diagnostics: An Expository Treatment of Outliers and Influential Cases." In John Fox and J. Scott Long, eds., Modern Methods of Data Analysis, Sage Publications, pp. 257-291.
Chatterjee, Samprit and F. Wiseman. "Use of Regression Diagnostics in Political Science Research." American Journal of Political Science 27: 601-613.
Mock, Carol and Herbert F. Weisberg. "Political Innumeracy: Encounters with Coincidence, Improbability, and Chance." American Journal of Political Science 36: 1023-1046.
Friday, April 15th
Problem set #4 assigned; problem set #3 due.
Gujarati, Chapter 11
Downs, George W. and David M. Rocke. (1979) "Interpreting Heteroscedasticity," American Journal of Political Science, v. 23, no. 4 (November) pp. 816-828.
Lemieux, Peter (1976) "Heteroscedasticity and Causal Inference in Political Research" Political Methodology 3: 287-316.
Friday, April 22nd Autocorrelation
Gujarati: Chapter 12
Berry and Feldman: Section 6
Hibbs, Douglas A. (1974) "Problems of Statistical Estimation and Causal Inference in Time-Series Regression Models," Sociological Methodology, pp. 252-308.