This preview shows pages 1–2. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: sentences. 6. Consider (3) again. Estimate the model and generate residuals ( u_hat i ) and predicted variables ( mort_hat i ). a. Compute the sample correlations between the residuals and log-tobacco, residuals and log-alcohol, and residuals and log-income. Present the correlations and their p-values. Explain why the result is trivial. b. Create a scatter plot of mort_hat (X-axis) and mort (Y-axis). Call it Figure 1 : Mortality and Predicted Mortality . Add labels, etc., using full names. Regress mort on mort_hat (i. .e. mort is the dependent or left-hand variable). Present the usual output in Table 4 . Use the scatter plot and Table 4 results to discuss how well (3) works as a model of mortality. Compare the R-squared in Tables 3 and 4 . What's going on? Can you prove why this must be so....
View Full Document
- Fall '08