G53.2127 : Quantitative Political Analysis II

Syllabus : December 18, 2000

Prerequisite: Students must have completed Quantitative Political Analysis I or the equivalent, or have approval of the instructor. Knowledge of rudimentary matrix algebra and calculus is assumed (we will review as necessary)..

Course Summary: This is a course in regression analysis. It begins with basic hypothesis testing using Ordinary Least Squares regression. In this course we will examine how to build more sophisticated models allowing us to test more complex hypotheses, and also learn more sophisticated statistical tests enabling us to proceed with analysis even when the Gauss-Markov assumptions are violated. The primary emphasis is on identifying statistical techniques appropriate to the question being examined, and correctly applying those techniques.

Course Requirements

There will be approximately 4 to 6 computer-based problem sets requiring students to analyze data with the techniques covered in class. You are free to work together on these problem sets, and are encouraged to do so! In addition, there will be an 18-25 page research paper on a topic chosen by the student - - with the professor's approval - - utilizing techniques covered in class. This paper can be a paper that the student is writing for another course, provided that it makes sufficient use of material covered in this course. If the paper is to be turned in to another class this needs to be made clear to both faculty members receiving the paper. Class attendance and participation are mandatory.

Your grade is computed as follows: Problem Sets: 30%; Paper: 70%.

Texts and Reading:

Required Books

Gujarati, Damodar N., Basic Econometrics, 3^rd edition, 1995, McGraw Hill.

Namboodiri, Krishman, Matrix Algebra: An Introduction, Sage Quantitative Applications in the Social Sciences, No. 38.

Recommended Books

Hamilton, Lawrence C. Statistics with Stata 5, 1997, Duxbury. ISBN: 0534265596.

Klepner, Daniel. Quick Calculus, 2^nd edition, 1985, John Wiley and Sons. ISBN: 0471827223. ($18 paperback).

Thompson, Silvanus Phillips. Calculus Made Easy, 1998, 3^rd edition, St. Martin's Press. ISBN: 0312114109. ($10 paperback).

Articles

Achen, Christopher H., ``Measuring Representation: Perils of the Correlation Coefficient'', American Journal of Political Science, 1977, Vol 21:805-815.

Beck, Nathaniel and Jonathan N. Katz., ``What to Do (and Not to Do) with Time-Series-Cross-Section Data," American Political Science Review, 89(1995):634-47.

Berk, Richard A., ``An Introduction to Sample Selection Bias in Sociological Data,'' American Sociological Review, June, 1983, Vol 48:386-398.

Green, Donald Philip , and Jonathan S. Krasno, ``Salvation for the Spendthrift Incumbent: Reestimating the Effects of Campaign Spending in House Elections,'' American Journal of Political Science, Nov, 1988, Vol 32:884-907.

Jacobson, Gary, ``The Effects of Campaign Spending in Congressional Elections,'' American Political Science Review, 1978, Vol 72:769-783.

Kiewiet, D. Roderick and McCubbins, ``Presidential Influence on Congressional Appropriations Decisions,'' American Journal of Political Science, August, 1988.

King, Gary, ``How Not To Lie With Statistics: Avoiding Common Mistakes in Quantitative Political Science, '' American Journal of Political Science, 1986, Vol 30:666-687.

King, Gary, ``Stochastic Variation: A Comment on Lewis-Beck and Skalaban's ``The R-Squared''', Political Analysis, 1990, Vol 2:185-200.

Kramer, Gerald H., ``The Ecological Fallacy Revisisted: Aggregate versus Individual-level Findings on Economics and Elections, and Sociotropic Voting'', American Political Science Review, 1983, Vol 77:92-111.

Nagler, J., ``Coding Style and Good Computing Practices'', PS: Political Science and Politics September, 1995 - Volume 28, No 3. Pages 488-492.

Robinson, W.S., ``Ecological Correlations and the Behavior of Individuals,'' American Sociological Review, 1950, Vol 15:351-357.

Wright, Gerald C., ``Linear Models for Evaluating Conditional Relationships,'', American Journal of Political Science, May, 1976, Vol 20:349-373.

Prof. Jonathan Nagler Spring, 2001

G53.2127 Quantitative Political Analysis II

Syllabus : December 18, 2000

Topics Covered

1  The basic linear model (Wks 1-4)

• matrix algebra
  • Read: Namboodiri, Chapter 1 (Skip Markov Chains).
  • notation
  • addition, subtraction, multiplication by a scaler
  • vector representation of a system of linear equations
  • inner products
  • matrix multiplication
  • the identity matrix

• The General Linear Model
  • parameters vs. estimates
  • a disturbance term
  • the interpretation of b

• OLS Estimation
  • Read: Guj: Chap 3.1-3.4
  • Read: Nagler: OLS Notes (online)
  • criteria for choosing [^(b)]
  • Properties of OLS (SANS assumptions!!!)
  • Gauss-Markov Theorem ® BLUE

• Using Statistical Software

  • Read: Nagler, PS: Political Science and Politics 1995.

• More OLS Estimation - Multiple Regression
  • Read: Guj: Chap 7
  • 3 Variable Models
  • Estimation and Variance of OLS Estimators
  • Interpretation of `partial regression coefficients'

• More OLS Estimation - F-Tests of Multiple Coefficients
  • Read: Guj: Chap 8
  • F-Tests
  • Increment of an additional variable

• R² vs. (std. error of [^Y])
  • Read: Achen (AJPS, 1977)
  • Read: King (Political Analysis, 1990)
  • Read: Guj: Chaps 3.5, 7.5, 7.8.

• The Race of the Variables

  • Read: King (AJPS, 1986), p. 669-678.

• More Matrix Algebra
  • Read: Namboodiri, Chapter 2 (Skip Echelon Matrices/Input-Output Analysis).
  • Read: Guj: Chap 9.1-9.3.
  • Elementary Operations
  • Inverse of a Matrix and how to Calculate it
  • Application of Inverse to System of Equations
  • Application of Inverse to Regression

• Common Violations of GM and Associated Symptoms
  • Read: Guj: Chapter 11
  • heteroscedasticity
  • misspecification

• Common Non-Violations of GM and Associated Symptoms
  • Read: Guj: Chapter 10
  • multicollinearity

• Common Violations of GM and Associated Symptoms

  • Read: Guj: Chapter 12
  • autocorrelation

2  Limited Dependent Variables (Wks 5-6)

• Read: Guj: Chapter 16
• Probit and Logit
• LHS variables bounded by 0-100

3  Model Specification: (Wks 7-9)

• Some Tests of Specification Errors
  • Read: Guj: Chap 13

• Omitted Variables
  • Read: Guj: Chap 7.7

• Multiple RHS Categorical Variables
  • Read: Guj: Chap 15.1-15.8

• Interactive Effects
  • Read: Wright (AJPS, 1976)

• Fixed Effects
• Switching Regime Models
  • Read: Kiewiet and McCubbins (AJPS, 1988)

• Leamer's Method of Bounds, ETC
  • Read: Guj: Chapter 14

4  Two-Stage Least Squares: IV Estimators (Wk 10)

• Read: Jacobson (APSR, 1978)
• Read: Greene and Krasno (AJPS, 1988)
• Read: Handout

5  Identification and Simultaneous Equation Models (Wks 11-12)

• Read: Guj: Chap 18, 19.

6  Time Series (Week 13)

• Lags

  • Read: Guj: Chapter 17

• Time-Series/Cross-Section

  • Read: Beck and Katz (APSR 1995)

7  Aggregation : individual level data vs. aggregate data (ecological inference) (Wk 14)

• Read: Robinson (ASR, 1950)
• Read: Kramer (APSR, 1983)
• Read: King (PUP, 1997), selected pages.

8  Factor Analysis