A researcher investigates the performance of real estate investment trust. She postulates that real estate investment performance is related to the stock market as well as the mortgage interest rate. Real estate investment trust is hypothesized to be positively related to the stock market and negatively related to mortgage interest rate. As part of the analysis, the following heteroscedasticity corrected regression model is estimated. The dependent variable is monthly return on REIT fund (RVNQ) and the independent variables are S&P500 index return (RSNP) and mortgage interest rate (mtg). Durbin-Watson statistics (DW) is also reported following the estimated table. RVNQt = 1 + 2*RSNPt + 3*mtgt + ut Coefficient Estimate Std.Error t value Pr(>|t|) Intercept -2.42407 2.24849 -1.0781 0.2843 RSNP 0.84412 0.13453 6.2747 1.75E-08 *** mtg 0.49018 0.58649 0.8358 0.4058 --- Signif. codes: 0 *** 0.001 ** 0.01 * 0.05 . 0.1 1 Residual standard error: 3.241 on 79 degrees of freedom Multiple R-squared: 0.5451, Adjusted R-squared: 0.5336

F-statistic: 47.33 on 2 and 79 DF, p-value: 3.078e-14 > dwtest(lm_RVNQ) Durbin-Watson test data: lm_RVNQ DW = 2.0571, p-value = 0.5633 alternative hypothesis: true autocorrelation is greater than 0 Test the following at 5% level. State the null and alternative hypotheses clearly. a) Test if RSNP and mtg have significant influence on RVNQ. b) RVNQ is claimed to be less risky than the stock market index. Test if that is true. c) Is there serial correlation problem? Test by using the DW result reported. Assume sample size of 80 observations.