Office Hours: 10-1 Wednesday

Office 913 PLC, 346-4868

This course focuses on survey of econometrics techniques with emphasis on applications in political science. Specific topics include introduction to basic statistical concepts, introduction to quantitative analysis, concepts and methods of empirical research, applied statistical data analysis in political science. Methods include descriptive statistics, bivariate correlation, and regression techniques. The spring part of the course will mostly focus on applied aspects of the methods studied during the winter term as well as on learning of using different types of statistical software. Approximately 40 percent of the time will be devoted to presentations and discussions. Designed to help students become informed users of quantitative methods in political science.

GRADING: There will be two take-home midterm exams in this course, and no final exam. The exams will be due on May. 2nd (30 points) and June. 10th (30 points). Each exam will be available at least two days before the due date. No makeups will be scheduled for any of the exams without a documented medical excuse. Aside from the midterm exams each student will be required to study a problem of his/her choice and make a presentation of the results (40 points). There will be a number of homeworks in this course. They will be graded as pass/fail. Generally a passing grade will be given to all turned in homeworks which show that some reasonable amount of work has been contributed, and at least an attempt was made to find correct answers. Each missing homework will result in a 5 point deduction from final score. Grades: A:85-100, B:70-84, C:50-69.

BOOKS: G. S. Maddala "Introduction to Econometri cs" - primary textbook

Peter Kennedy "A guide to Econometrics" - secondary textbook

Apr. 1-4 Multiple Regression. A Model with Two Explanatory Variables. Statistical Inference in the Multiple Regression Model. Interpretation of Regression Coefficients Reading Maddala 128-143.

Apr 7-11. Multiple Regression. Partial Correlations. Relationships among the explanatory variables. Prediction in the multiple regression model. Analysis of variance and tests of hypothesis. Omission of relevant variables and inclusion of irrelevant variables. Maddala 146-161.

Apr 14-18. Introduction into different types of statistical software: Statistica, SPSS, SST, Stata. Examples of multiple regression analysis using statistical software packages. Heteroskedasticity: Detection, Consequences and Solutions. Reading: Maddala 201-220.

Apr 21-25. Autocorrelation: Durbin-Watson test, Estimation in levels versus first differences. Estimation procedures with autocorrelated errors. Effect of AR(1). Reading: Maddala 229-264.

Apr 28-May 2. Multicollinearity: measures, problems and solutions. Principle component regression. Ridge regression. Dropping variables. Illustrative examples. Midterm Exam (May 2nd). Reading Maddala 269-292.

May 5-9. Discrete Choice Analysis. Dummy Variables: intercept, slope ect. The linear probability model and the linear discriminant function. The Probit and Logit models. Truncated variables: the Tobit model.

May 12-16. Simultaneous Equations Models: Endogenous and Exogenous Variables. The identification problem. Necessary and Sufficient condition for Identification. Methods of estimation. Reading: Maddala 356-377.

May. 19-23. Errors in Variables and Specification Testing. Reverse Regression. Proxy variables. Diagnostic Test Based on Least Squares Residuals. Some other types of residuals. Selection of Regressors. Reading Maddala 447-500.

May. 26- 30. Introduction to Time-Series Analysis. Stationary and Nonstationary Time Series. Some useful models for time series : Maddala 525-549.

May. 6-10 Review, Midterm Exam (May 10th).