Question 2 [4+2+3+2+4=15] (Chapter 5, Section 5.2) The output be- low was obtained from fitting a three - covariate linear model, Yi = Bot>Bjrij + eis j=1 to observed responses: covariate est.coef. st.err. sig. 0.5 0.02 1.0 0.25 0.00 0.4 0.4 0.33 RSS = 100; n = 30; R' = 0.9 where of is short for "degrees of freedom", est.coef. denotes the least squares estimates of 8;, j = 1, 2, 3, and sig. denotes the p-value (two- sided) of the t-statistic for testing the hypothesis that the coefficient of the respective covariate is zero. 2.1 Find an unbiased estimate of the error variance, o?. 2.2 Find the two possible numerical values of B1. 2.3 Find a 95% confidence interval for 2, the coefficient of X2 in the model. 2.5 Test at the 5% level the hypothesis Ho : B1 = B2 = B3 = 0. NOTE: Here you could use either the Analysis of Variance method (page 135) or the Partial F-test method (page 137) 2.4 Find the regression sum of squares, SSreg, for these data