B. Dan Wood
Political Science 606.600
Dynamic Analysis with Quantitative Methods
Fall Semester, 2007

Location: 2064 Bush Academic Building West

Office: 2098 Bush Academic Building West

Time: 7:00-9:50 p.m. Tuesday

Office Hours: 6:00-7:00 p.m. Tuesday

Phone: 845-1610

Email: bdanwood@polisci.tamu.edu

 

Web: Texas A&M Political Science Department – B. Dan Wood

Purpose- This course considers statistical techniques to evaluate social processes occurring through time. The early focus is on econometric regression methods, followed by a Box-Jenkins perspective, with later attention given to more recently developed approaches such as Error Correction models, multivariate time series (VAR), and pooled time series cross sections. We will also look at time varying parameter models including ARCH, regime switching, state space, and flexible least squares.  The emphasis throughout the course will be on application, rather than on statistical theory. However, the focus of most lectures will be statistical theory.

It is assumed that you bring to the course a background in high school level algebra and statistics up to and including regression. You should also have some experience with microcomputers. A primary statistical package for the class will be RATS for Windows (WINRATS). We will also be working in R. However, you need not be proficient in RATS or R before entering the course since sample programs will be available.

Grades- The course grade will be determined by your performance on the homework assignments and class presentations (33%), a research project in which you apply one or more of the tools presented in this course (33%), and a final examination (33%). Homework assignments are due on the date listed on the attached reading list. I will call on each of you at some point to discuss aspects of the homework assignment. Homework is due the next week after the assignment on the following course outline. The research project is due on Tuesday, December 4th.  Late homework or research projects will not be accepted. The final examination will be take home and due one week after the end of classes.

Book Requirements-

Enders, Walter. Applied Econometric Time Series, Second Edition. Wiley.
Enders, Walter. RATS Handbook for Econometric Time Series. Wiley.
Shumway, Robert H. and David S. Stoffer. Time Series and Its Applications: with R Examples. Springer
Pfaff, Bernard.
Analysis of Integrated and Cointegrated Time Series with R. Springer.
Doan, Thomas A. RATS User’s Manual, Version 6. Evanston, IL: Estima.
McCleary, Richard and Richard Hay. 1980. Applied Time Series Analysis. Sage Publications. This is out of print. I will supply a copy for Xeroxing.
Selected illustrative articles as recommended in the attached course outline.

The RATS Handbook contains a disk with some of the data you will need for completing the exercises.  Click here for the data for other assignments. 

The Americans with Disabilities Act (ADA) is a federal anti-discrimination statute that provides comprehensive civil rights protection for persons with disabilities. Among other things, this legislation requires that all students with disabilities be guaranteed a learning environment that provides for reasonable accommodation of their disabilities. If you believe you have a disability requiring an accommodation, please contact the Office of Support Services for Students with Disabilities in Room 126 of the Student Services Building. The phone number is 845-1637.

Topics and Readings

r=required reading
n=recommended, but not required
a=assignment

1. Introductory overview: administrative matters primarily. During this session we will be positioning Time Series Analysis within the larger framework of statistical analysis and research design. Additionally, we will be introducing the computer software for the course.

r--. Read RATS Users Manual, chapters 1 through 3.
r--. Read Shumway, chapter1.
r--. Read Greene, William H.  2003. Econometric Analysis: 5th Edition. Chapter 12.
a--. Do Enders Handbook, chapter 1, using WINRATS.  Do R-code examples in Shumway, chapter 1. Be prepared to discuss the materials. 

2. A survey of elementary regression time series methods: autocorrelation. (The following reading assignments should be completed over the next three weeks.)

r-- Read Granger, C.W. J. and P. Newbold. 1974. Spurious Regressions in Econometrics. Journal of Econometrics. 2: 111-120.
r-- Read Enders, Chapters 1, 2, and 4.  You can just skim the differential equation stuff in chapter 1 if it seems too difficult.
r- Read Shumway, chapters 2 and 3.
r-- Read Pfaff, chapters 1, 2,4, and 5.
r—Look at McCleary and Hay, Chapters 1&2, pp. 17-139 for a simplified treatment.
n-- Chatfield, C. 1979. Inverse Autocorrelations. Journal of the Royal Statistical Society. 142: 363-377.
n-- Schwartz, G. 1978. Estimating the Dimension of a Model. Annals of Statistics. 6: 461-464.
n-- Akaike, H. 1974. A New Look at Model Identification. IEEE Transaction On Automatic Control. AC-19: 716-723.
a-- The Ostrom data (see link above) contains US and USSR defense spending for 1967-88.  Use RATS to estimate a model of US defense spending over this period as a function of USSR defense spending.  Test for residual autocorrelation.  See RATS Users Guide section 6.5.  Estimate a series of models that consider the serial correlation.  (Hint: you might consider the LINREG procedure with robust standard errors; you might also consider the AR1 procedure with various options; you might also consider a lagged dependent variable or various transformations of the data.)  Diagnose the adequacy of your final model.  Critique the model in terms of violations of the assumptions of regression.
a – Do Enders Handbook, chapter 2.
a -- Do R-code examples in Shumway, chapter 2.

3. Time Series methods: introduction to ARIMA models with a focus on identification and stationarity issues.

a--. Using either RATS or R, graph the German Immigration, IBM B, and Mine Injuries Series. Use RATS or R to identify the different ARIMA components of these series. Make sure that you do appropriate transformations of these series to assure level and variance stationarity. In RATS use @DFUNIT, @PPUNIT, and @BAYESTST to do formal hypothesis tests for the level stationarity of these and the transformed series.  Also, visit Estima’s web site and explore some of the various alternative procedures for exploring whether there is a unit root.  In particular, download and use @URADF, @KPSS and @VRATIO to explore the presence of a unit root in these series. If using R, then consider the procedures in Shumway, chapter 3 and/or Pfaff, chapters 1, 2, and 4.
a--.Do R-code examples in Pfaff, chapters 2 and 4.
a--. D R-code examples in Shumway, chapter 3.
a -- Do R-code examples in Pfaff, chapter 1.

4. Building univariate noise models: estimation, diagnosis, and metadiagnosis.

a--. Use RATS or R to identify and estimate a univariate ARIMA model for the Sutter County Workforce and Boston Armed Robery time series. Replicate (if possible) the results reported by McCleary and Hay for these two series on pp. 104-121. Offer a written critique of the models they report there.

5. Building univariate noise models: estimation, diagnosis, and metadiagnosis (continued).

a--. Use RATS or R to identify and estimate a univariate ARIMA model for the Swedish Harvest Index and Hyde Park Purse Snatchings series. Replicate (if possible) the results reported by McCleary and Hay for these two series on pp. 121-132. Offer a written critique of the models they report there.

6. ARIMA impact assessment: transfer function approaches. (Read the following over the next two weeks.)

r--. Read McCleary and Hay, Chapters 3 and 4, pp. 141-228.
r--. Read Enders, Chapter 5, pp. 240-247.
r--. Read Shumwaya, Chapter 5, pp. 295-301.
r--. Read Wood, B. Dan. 1988. Principals, Bureaucrats, and Responsiveness. American Political Science Review. 82: 213-234.
n--. Read Box, G.E.P. and G.C. Tiao. 1975. Intervention Analysis with Applications to Economic and Environmental Problems. Journal of the American Statistical Association. 70: pp. 70-79.
n--. Read Hibbs, Douglas A. Jr. 1977. Political Parties and Macroeconomic Policy. American Political Science Review. 71: 1467-1479.
n--. Read Moe, Terry M. 1982. Regulatory Performance and Presidential Administration. American Journal of Political Science. 26: 197-224.
n--. Read Rasler, Karen. 1986. War, Accommodation, and Violence in the United States, 1890-1970. American Political Science Review. 80: 921-945.
n--. Read Hibbs, Douglas A. Jr. 1977. On Analyzing the Effects of Policy Interventions: Box-Jenkins and Box-Tiao vs. Structural Equation Models. Sociological Methodology: 1977.
n--. Read Oppenheimer, Bruce I., James A. Stimson, and Richard W. Waterman. 1986. Interpreting U.S. Congressional Elections: The Exposure Thesis. Legislative Studies Quarterly. 11: 227-247.

a--. Use RATS to build intervention models for the Directory Assistance and Schizophrenic Perceptual Speed series. Replicate McCleary and Hay's analyses of these series (if possible) as reported on pp. 145-164. Offer a written critique of their results. Additionally, use RATS to build intervention models for the Sutter County Workforce, Minneapolis Public Drunkenness, and Hyde Park Purse Snatchings series. Replicate McCleary and Hay's analyses of these series (if possible) as reported on pp. 164-191, 199-201. Offer a written critique of their results.
an--. Use RATS to build an intervention model for the Suicides series. Replicate McCleary and Hay's analyses of this series (if possible) as reported on pp. 191-199.

7. Transfer functions and causality. (Read the following over the next two weeks.)

r--. Read Enders, Chapter 5, pp. 247-264.
r--. Read McCleary and Hay, Chapter 5.
r--. Reread Shumway, chapter 5, pp. 295-301
r--. Read Norporth, Helmut. Transfer Function Analysis. New Tools for Social Scientists. W.D. Berry and Michael Lewis-Beck eds. Sage Publications: Beverly Hills.
r--. Read Freeman, John R. 1983. Granger Causality and Time Series Analysis of Political Relationships. American Journal of Political Science. 27: 327-358.
n--. Sheehan, Richard G. and Robin Grieves. Sunspots and Cycles: A Test Of Causation. Southern Economic Journal. 1982: 775-77.
n--. Read Carmines, Edward G. and James A. Stimson. 1986. On the Structure and Sequence of Issue Evolution. American Political Science Review. 80: 901-920.
a--. Use RATS to build a transfer function causal model for the IBM Paris-New York series. Replicate McCleary and Hay's analyses of these series (if possible) as reported on pp. 236-243. Offer a written critique of their results. Additionally, use RATS to build a bivariate transfer function causal model for the Swedish Population and Swedish Harvest series and a multivariate transfer function causal model for the Swedish Population, Swedish harvest, and Swedish Fertility series. Replicate McCleary and Hay's analyses of these series (if possible) as reported on pp. 243-270. Offer a written critique of their results.

8. Bridging the gap between ARIMA and regression time series models.

r--. Read Harvey, A.C. 1990. The Econometric Analysis of Time Series, Chapter 7. Cambridge: MIT Press.
r--. Read Beck, Nathaniel. 1991. Comparing Dynamic Specifications: The Case of Presidential Approval. Political Analysis. Ann Arbor, MI: University of Michigan Press.
r--. Lewis-Beck, Michael S. 1986. Interrupted Time Series. New Tools For Social Scientists. William D. Berry and Michael Lewis-Beck, eds. Beverly Hills: Sage Publications.
n--. Wonnacott, Ronald J. and Thomas H. Wonnacott. 1979. Econometrics. pp. 212-233.
n--. Caldiera, Gregory A. 1987. Public Opinion and the U.S. Supreme Court. American Political Science Review. 81: 1139- 1154.
n--. Wood, B. Dan. 1990. Does Politics Make a Difference at the EEOC? American Journal of Political Science. 34: 503-30.
n--. Read Beck, Nathaniel. 1985. Estimating Dynamic Models Is Not Merely A Matter Of Technique. Political Methodology. 11: 71- 89.
n--. Read Monroe, Kristen R. 1981. Presidential Popularity: An Almon Distributed Lag Model. Political Methodology. 7: 43-69.
a--. Replicate the analyses reported in Beck (1991) Tables 1 through Table 4 (if possible). Discuss how all of these models relate to one another. Which do you think is a best model?

9. Bridging the gap: Vector Autoregressive and VARMA Models

r--. Read Enders, Chapter 5, pp. 264-311.
r--. Read Shumway, chapter 5, pp. 302-319.
r--. Read Judge, George G., W. E. Griffiths, R. Carter Hill, Helmut Lutkepohl, and Tsoung-Chao Lee. Introduction to the Theory and Practice of Econometrics. Chapter 18
r--.
Read Maddala, G.S. 1992. Introduction to Econometrics: Second Edition. Chapter 14.
r--. Read Freeman, John R., John T. Williams, and Tse-min Lin. Vector Autoregression and the Study of Politics. American Journal of Political Science. 33: 842-877.
r--. Read Freeman, John R, Hauser, Kellstedt, and Williams. 1998.  Long-Memoried Processes, Unit Roots, and Causal Inference in Political Science. American Journal of Political Science.  42:1289-1327.
n--. Williams, John T. 1990. The Political Manipulation of Macroeconomic Policy. American Political Science Review. 84: 767-96.
n--. Read Huang, Chi. 1989. Determining the Lag Order of a Vector Autoregressive Process: Some Guidelines for Political Studies. Texas A&M University Working Paper.
a--. Do Enders Handbook, Chapter 5.  Visit the Estima web site to see VAR tools.  Note: You can also implement VAR in R using the contributed package dse1. Not as convenient though.

10. Bridging the gap: cointegration and error correction models

r--. Enders, Chapter 6.
r--. Pfaff, Chapters 3, 6, and 7.
r--. Engle, R.F. and C.W.J. Granger. 1991. Introduction. Long Run Economic Relationships: Readings in Cointegration. New York: Oxford University Press.
r--. Greene, William H. 2000. Econometric Analysis: Fourth Edition. Section 18.4.
n--. Ostrom, Charles W., Jr. and Renee M. Smith. 1994. Cointegration and Error Correction in Multiple Time Series Analysis: Presidential Approval and the Quality of Life Equilibrium Hypothesis. Political Analysis. Volume 4: 127-184.
n--. Durr, Robert. 1994. Political Analysis. Volume 4: 185-228.
n--. Williams, John, 1994. Political Analysis. Volume 4: 229-236.
n--. Beck, Nathaniel. 1994. Political Analysis. Volume 4: 237-248.
n—De Boef, Suzanna and Jim Granato.  2000.  Testing for Cointegrating Relationships with Near Integrated Data.  Political Analysis.  8: 99-117.
n--. Engle, R. F. and C.W.J. Granger. 1987. Cointegration and Error Correction: Representation, Estimation, and Testing. Econometrica. 55: 251-276.
n--. Engle, R. F. and B. Sam Yoo. 1987. Cointegrated Economic Time Series: An Overview with New Results. Paper presented at the European Meeting of the Econometric Society in Copenhagen, August 1987.
n--. MacKinnon, James G. 1991. Critical Values for Cointegration Tests. Long Run Economic Relationships: Readings in Cointegration. New York: Oxford University Press.
a--. Do Enders Handbook, Chapter 6.  Visit the Estima website to obtain various canned procedures for exploring the presence of cointegration. Do the R-code examples in Pfaff, chapters 3, 6, and 7.

11. Unit Roots, and Integrated Data Revisited: Fractional Integration, and Near Unit Roots

r--. Read Greene.  2003. Econometric Analysis: Fifth Edition.  Section 20.3.5.
r--. Read Shumway, chapter 5, pp. 271-279.
r--. Read Pfaff, chapter 2, pp. 30-36.
r--. Read Box-Steffensmeier, Janet and Renee Smith.  1998.  Investigating Political Dynamics Using Fractional Integration Methods. American Journal of Political Science. 42: 661-689.
r--. Read Box-Steffensmeier, Janet, Kathleen Knight, and Lee Sigelman.  1998.  The Interplay of Macropartisanship and Macroideology: A Time Series Analysis.  Journal of Politics.  60: 1031-1049.
r--. Read DeBoef, Suzanna and James Granato.  1997. Near Integration and the Analysis of Political Relationships. American Journal of Political Science. 41: 619-640.
r--. Read DeBoef, Suzanna and James Granato, Overdifferencing: Implications for the Time Series Study of Political Relationships. Working paper.
a--. Use either RATS or R to replicate (if possible) the results in Table 5 of Box-Steffensmeier and Smith.  Visit the Estima website to obtain various canned procedures for dealing with Fractional Integration. Also, do the R-code examples in Pfaff, chapter 2, pp. 30-36.

12. Time Varying Parameters: ARCH, Regime Switching, and Other Time Varying Parameter Methods.

r--. Read Enders, Chapter 3.
r--. Read Shumway, chapter 5, pp280-295.
n--.  Read Sayrs.  1993. 
The Long Cycle in International Relations: A Markov Specification

International Studies Quarterly, Vol. 37, No. 2. 215-237.

n--. Read Freeman and Houser.  1998.  A Computable Equilibrium Model for the Study of Political Economy American Journal of Political Science, Vol. 42, No. 2. 628-660.
n--. Read Wood.  2000.  Weak Theories and Parameter Instability: Using Flexible Least Squares to Take Time-Varying Relationships Seriously. American Journal of Political Science.  44: 603-618.
n--Read Wood.  2000. The Federal Balanced Budget Force: Modeling Variations from 1904-1996. Journal of Politics.  62:817-845.
n—Wood.  2002. 
The Time Varying Effect of Public Approval on Presidential Success in Congress.  Journal of Politics 2003, with Jon Bond and Richard Fleischer.
a--. Do Enders Handbook, Chapter 3.  Visit the Estima web site to see extra procedures for estimating ARCH and Regime Switching models. Do R-code examples in Shumway, sections 5.3 and 5.4.

13. Regression With Pooled Data. (These readings and assignments cover the last two weeks.)

r--. Read Pindyck, Robert S. and Daniel Rubinfeld, 1981. Pooling of Cross-Section and Time Series Data. Econometric Models and Economic Forecasts. McGraw-Hill: New York. pp. 252-261.
r--. Read Hsiao, Cheng. 1986. Analysis of Panel Data, Chapters 1&2. Cambridge: Cambridge University Press.
r--. Read Stimson, James A. 1985. Regression in Space and Time: A Statistical Essay. American Journal of Political Science. 29: 914-947.
r--. Read and study Judge, George G., W. E. Griffiths, R. Carter Hill, Helmut Lutkepohl, and Tsoung-Chao Lee. The Theory and Practice of Econometrics. Table 13.1
r--.
Read Sayrs, Lois. Pooled Time Series Analysis. Sage Publications: Beverly Hills.
r--. Read Greene, William H. 2003. Econometric Analysis: Fifth  Edition. Chapter 13.                                                                                                                                                           
a--. Use the data in Greene, 2003, Appendix Table F7.1 to replicate the applicable analyses in Greene, chapter 13.  Enter the data in RATSDATA. Then use RATS's Panel procedure to replicate (if possible) the results reported in Greene Examples 13.2 and 13.5.

14. Regression With Pooled Data (cont.)

r--. Read Greene, William H. 2003. Econometric Analysis: Fifth Edition. Chapter 13.
r--. Read Beck, Nathaniel and Jonathan Katz. 1995. What to do (and Not to Do) with Time Series-Cross Section Data in Comparative Politics. American Political Science Review. 89: 634-47.
a--. Use Greene's data in Table F13.1 to explore pooled time series cross section estimation. Enter the data. Then use RATS's Panel procedure to replicate (if possible) the results reported in Greene Chapter 13.9.7.