9. You believe there is a relationship between book-to-market ratios and subsequent returns. The output from a cross- sectional regression and a graph of the actual and predicted relationship between the book-to-market ratio and return are shown below. Results from Regressing Returns on the Book-to-Market Ratio Coefficient 12.0130 -9.2209 Standard Error 3.5464 8.4454 t-Statistic 3.3874 -1.0918 DF MSS F Intercept Book value Market value) ANOVA Regression Residual Total Significance F 0.2831 1 1.1921 SS 154.9866 4162.1895 4317.1761 154.9866 130.0684 32 33 Residual standard error R-squared Observations 11.4048 0.0359 34 Return (%) Return versus Book-to-Market Ratio: Actual and Predicted 48 PRED RETURN 40 32 24 4 16 O O OOO DOCT mo 8 DO 8 0 O 0 O o -8 0,00 0.16 0.32 0.48 0.64 0.80 0.96 1.12 Book-to-Market Ratio A. You are concerned with model specification problems and regression assumption violations. Focusing on assumption violations, discuss symptoms of conditional heteroskedasticity based on the graph of the actual and predicted relationship. B. Describe in detail how you could formally test for conditional heteroskedasticity in this regression. C. Describe a recommended method for correcting for conditional heteroskedasticity. 9. You believe there is a relationship between book-to-market ratios and subsequent returns. The output from a cross- sectional regression and a graph of the actual and predicted relationship between the book-to-market ratio and return are shown below. Results from Regressing Returns on the Book-to-Market Ratio Coefficient 12.0130 -9.2209 Standard Error 3.5464 8.4454 t-Statistic 3.3874 -1.0918 DF MSS F Intercept Book value Market value) ANOVA Regression Residual Total Significance F 0.2831 1 1.1921 SS 154.9866 4162.1895 4317.1761 154.9866 130.0684 32 33 Residual standard error R-squared Observations 11.4048 0.0359 34 Return (%) Return versus Book-to-Market Ratio: Actual and Predicted 48 PRED RETURN 40 32 24 4 16 O O OOO DOCT mo 8 DO 8 0 O 0 O o -8 0,00 0.16 0.32 0.48 0.64 0.80 0.96 1.12 Book-to-Market Ratio A. You are concerned with model specification problems and regression assumption violations. Focusing on assumption violations, discuss symptoms of conditional heteroskedasticity based on the graph of the actual and predicted relationship. B. Describe in detail how you could formally test for conditional heteroskedasticity in this regression. C. Describe a recommended method for correcting for conditional heteroskedasticity
