Date Topic Reading Handout (including data and do-files) Homework (if any) Weeks 1-2 OLS in matrix form and MATA. Replication & do-files. LaTeX. CLARIFY and simulation. Nagler (1995), King (1995), King, Tomz & Wittenberg (2000), Greene (chp. 2-3) Do-files, Simulation and CLARIFY, notes on LaTeX, LaTeX example, OLS in matrix form. Homework 1: LaTeX and replication issues)
Homework 2: OLS in matrix form.Week 3 MLE - theory and mechanics. Properties of MLE. Gridsearches. Regression by MLE. King (2001, chp. 1-4), Greene (chp. 17, appendix E.6), Long (1997, chp. 4), Gould et al. (2003, chp 1- skim, 2-3) MLE - theory and mechanics Homework 3: Introduction to MLE Week 4 Binary response models - logit, probit, scobit, heteroskedastic probit. Long (1997, chp. 3), Greene (2003, 21.1-21.4), Wooldridge (2002, chp. 15.1-15.7), Nagler (1994), Herron (1999) + an applied article. Binary response models Homework 4: Binary response models Week 5 Ordered response models - ordered probit and logit, generalized ordered logit, heteroskedastic ordered probit. Long (1997, chp. 5), Greene (2003, 21.8), Wooldridge (2002, chp. 15.10) + an applied article. Ordered response models Homework 5: Ordered response models Week 6 Unordered response models - multinomial logit, conditional logit, nested logit, multinomial probit, mixed logit. Long (1997, chp. 6), Greene (2003, 21.7), Wooldridge (2002, 15.9), Alvarez & Nagler (1998), Garrett (2001) + 2 of the 3 applied pieces. Unordered response models Homework 6: Unordered response models Week 7 Event count models - poisson models, negative binomial models, generalized event count models, hurdle models, zero-inflated models. Long (1997, chp. 8), Greene (2003, 21.9), King (1989), Zorn (1998) + 1 of the two applied pieces. Event count models Homework 7: Event count models Weeks 8 Continuous time duration models: Univariate and bivariate analyses using Kaplan-Meier plots. Parametric models (exponential, weibull, log-logistic, generalized gamma etc.). Proportional hazards versus accelerated failure-time approaches. Estimation and interpretation - hazard rates, expected duration etc. Model selection. Cox proportional hazards model. Estimation and interpretation. Proportional hazards assumption. Schoenfeld, martingale, Cox-Snell residuals. Box-Steffensmeier and Jones (2004, chp. 1-4, 7-8), Greene (2003, 22.5), Wooldridge (2002, chp. 20), Box-Steffensmeier and Zorn (2001), King, Alt, Laver and Burns (1990) + one other applied piece. Continuous Time Duration Models, Brad Jones notes, Chris Zorn's notes, Neal Beck's notes, TVC example using Cox Homework 8: Continuous Time Duration Models Week 9 Discrete time duration models: Connection to TSCS models with binary dependent variable. Logit, probit, cloglog. Time dependence - robust clustered standard errors, temporal dummies, and cubic splines. Ongoing events and second spells. Time-varying covariates. Markov transition models. Box-Steffensmeier and Jones (2004, chp.5), Beck, Katz, and Tucker (1998), Beck et al. (2002), Alt, King and Signorino (2000), Oneal and Russett (1999) and one other applied piece. Discrete Time Duration Models, Brad Jones' notes
Homework 9: Discrete Time Duration Models
Week 10 Duration models cont'd.: Competing risk models, heterogeneity, frailty models, repeated events. Box-Steffensmeier and Jones (2004, chp.9-10), Zorn (2000), Box-Steffensmeier and Zorn (2002), Cleves (1999), Diermeier and Stevenson (1999) More Duration Notes, Chris Zorn's notes, Brad Jones' notes Homework 10: More on Duration Models Week 11 Tobit models: Truncated and censored data. Estimation and interpretation - multiple marginal effects and first differences. Comparing tobit, probit and OLS. Long (1997, chp. 7), Wooldridge (2002, 16.1-16.6), Greene (2003, 22.1-22.3), Sigelman and Zeng (1999), Golder (2003) Tobit Model Homework 11: Tobit Models Week 12 Selection models: Heckman model (two-step and MLE). Bivariate probit, bivariate probit with partial observability, and bivariate probit with sample selection. Greene (2003, 22.4), Heckman (1979), Sartori (2003), Dubin and Rivers (1989/1990), Berinsky (2004, chp. 3-5), Przeworski and Vreeland (2002). Selection Models Homework 12: Selection Models Week 13 Matching models: Rosenbaum and Rubin (1983), Ho, Imai, King and Stuart (2006), King and Zeng (2006), King and Zeng (2006a), Barabas (2004), Imai (2005), Epstein, Ho, King and Segal (2005). Matching No homework this week. Weeks 14-15 Time Series - Stationary and non-stationary data. Durbin-Watson and LM tests for serial correlation. Cochrane-Orcutt and Prais-Winsten procedures. Lags - finite distributed, ADL etc. Error correction models. Impulse and unit response functions. AR and MA error processes. Tests for stationarity and unit roots. Drifts and trends. Cointegration. Structural stability. Spurious regression. Greene (2003, Chp. 19-20), Gujarati (2003, Chp. 12, 17), Keele and Kelly (2006), Keele and DeBoef (2005), Murray (1994), Charemza and Deadman (1997, Chp. 3-5) + 2 applied pieces Homework 14: Time Series Weeks 16 Longitudinal data - (i) Panel vs TSCS data. Heterogeneity - fixed vs random effects. GLS, FGLS, Parks Methods, PCSEs etc. Lags. Hurwicz bias. Greene (2003, 13.1-13.4), Beck and Katz (1995), Hsiao (2003, chp. 1, 3).