Unformatted text preview: PADP 8120 Fertig Final Exam Review Topics Note: The Final is cumulative but the emphasis of the exam will be on these topics. Spring 2011 UGA Ordinary least squares (OLS) Minimizes the sum of squared deviations Bivariate formulas for intercept and slope Assumptions underlying OLS Multiple regression (purpose, implementation, and interpretation) Dependent vs. independent variables Error term Standard error of coefficient estimated by OLS (formula) Confidence interval for coefficient estimated by OLS (formula) Hypothesis testing with respect to estimated coefficients Correlation Residual Dummy variables Omitting reference category when including categorical dummy variables Interaction effects (purpose, implementation, and interpretation) R-squared (formula, purpose, interpretation) How to handle non-linear relationships in OLS (polynomials, log transformation) Heteroskedasticity (how to tell, what to do) Robust standard errors Outliers Leverage Multicollinearity Variance Inflation Factor Linear Probability Model Logistic regression (logit) Log odds equation Uses maximum likelihood estimation (MLE) Transforming coefficients to interpret as odds ratio Transforming coefficients to interpret as probabilities Interpreting coefficients when there are multiple independent variables Interpreting interaction effects Likelihood-ratio (LR) test (formula, purpose, implementation, and interpretation) Stata commands Dfbeta Set seed Lowess Reg Rvfplot Pwcorr Reg vars, vce(robust) Lfit Collin Predict var, residual Logit Predict var, rstudent Predict ...
View Full Document
- Summer '11
- Regression Analysis, OLS, Lfit Collin, residual Logit