Financial_Modeling_Midterm_13

0001 rsq037 rmse15065 summary of fit rsquare rsquare

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: E) using all the other variables has been fitted (the marital status was split in the correspondent set of dummy variables, i.e. MARSTAT_0 equals 1 for those separated/divorced/widowed and 0 otherwise, MARSTAT_1 equals 1 for those married and 0 otherwise, MARSTAT_2 equals 1 for those living with partner and 0 otherwise). The JMP outputs (REPORT 1 and REPORT 2) for this model are reported in the next pages. a) Using the information in REPORT 1, which one is the most important variable in predicting the log-transformed face amount? Why? Report also the interpretation of the MARSTAT_1 estimated coefficient. b) Predict the face amount for a family with INCOME = 70000, EDUCATION = 15, NUMHH = 4, MARITAL STATUS = “married”, AGE = 40, GENDER = “male”. c) Would you recommend to the insurance company using this model for making predictions about the amount insured for the potential customers? Provide a detailed explanation for your answer. d) REPORT 2 contains the results for the F test on a subset of the model coefficients. Explain what it does test for and what conclusions you can get about it from the output. 7 REPORT 1 Response LN_FACE Whole Model LN_FACE Actual Actual by Predicted Plot 17 16 15 14 13 12 11 10 9 8 7 7 8 9 10 11 12 13 14 15 16 17 LN_FACE Predicted P<.0001 RSq=0.37 RMSE=1.5065 Summary of Fit RSquare RSquare Adj Root Mean Square Error Mean of Response Observations (or Sum Wgts) 0.368028 0.351459 1.506535 11.99029 275 Analysis of Variance Source Model Error C. Total DF 7 267 274 Sum of Squares Mean Square F Ratio 352.90025 50.4143 22.2124 605.99610 2.2696 Prob > F 958.89635 <.0001* Parameter Estimates Term Intercept LN_INCOME EDUCATION NUMHH MARSTAT_ 0 MARSTAT_ 1 AGE GENDER Estimate 2.2136787 0.4360532 0.213627 0.2342311 0.6729936 0.8521004 -0.005916 0.714108 Std Error 1.067733 0.079038 0.038708 0.074229 0.575681 0.497643 0.008036 0.372637 t Ratio Prob>|t| Lower 95% Upper 95% 2.0733 0.0391* 0.1114306 4.3159269 5.5170 <.0001* 0.2804365 0.59167 5.5190 <.0001* 0.1374162 0.2898378 3.1555 0.0018* 0.0880823 0.3803799 1.1690 0.2434 -0.460459 1.8064462 1.7123 0.0880 -0.127703 1.831904 -0.7362 0.4623 -0.021738 0.0099062 1.9164 0.0564 -0.019572 1.4477879 Residual by Predicted Plot 4 LN_FACE Residual 2 0 -2 -4 -6 7 8 9 10 11 12 13 LN_FACE Predicted 14 15 16 8 17 VIF . 1.2647957 1.1754328 1.4822372 6.5978817 5.5294598 1.1788516 1.9591079 REPORT 2 Response LN_FACE Custom Test F Test Parameter Intercept LN_INCOME EDUCATION NUMHH MARSTA...
View Full Document

This note was uploaded on 03/27/2014 for the course FINANCE v taught by Professor G during the Fall '14 term at Università Bocconi.

Ask a homework question - tutors are online