Vous êtes sur la page 1sur 1

Quiz Questions Chapter 7: Assessing Studies Based on Multiple Regression

1. Provide a definition for each of the Key Terms at the end of the chapter. 2. Why is internal validity an important criterion for evaluating an econometric study? 3. Why is external validity an important criterion for evaluating an econometric study? 4. Key Concept 5.4 lists four least squares assumptions for the multiple regression model. Which of these assumptions is/are likely fail in the following circumstances? a. b. c. d. e. There is misspecification of the regressions functional form. The regressors are measured with error. Data are not available when the dependent variable falls in a certain range. There is reverse causation flowing from Y to one of the regressors. The regression error terms are correlated across observations.

5. Worried that his regression may suffer from omitted variable bias, a researcher includes an additional regressor. The t-statistic on the coefficient for this regressor is t = 1.2. Should the researcher keep this variable in the regression or discard it? Explain.

Multiple Choice 1 Including an additional regressor in a regression a. always leads to more accurate estimates of regression coefficients. b. always reduces the sum of squared residuals. c. is advisable only if the variables t statistic is larger than 1.96 in absolute value. d. increases imperfect multicollinearity and is not advisable. 2. Simultaneous causality bias a. leads to regression coefficients that are too small in absolute value. b. arises because at least one of the regressors is correlated with the regression error. c. can be avoided by using accurate measurements of Y and X. d. is uncommon in economics, because causality is usually from one variable to another. 3. If regression error terms are correlated with one another, then a. the least squares estimators are biased. b. the least squares estimators are inconsistent. c. t-statistics will not be distributed as standard normal variables in large samples. d. the regression will suffer from perfect multicollinearity.

Vous aimerez peut-être aussi