ANOVA for Regression We're trying to fit a simple linear regression model for the the whole population:Y=Bo + B1X + E. (Assume & are are independent and normally distributed with constant variance). We draw a random sample of n=5 from the population and get a sample correlation r = 0.5. Compute the 4 test statistics for testing the null Ho: slope=0 in the population. (This can be written as Ho: B1=0. It's also the same as testing whether population =0) (Round your final answers to 2 decimal places, but don't round during intermediate steps.) a. First compute: R2 = 1-R2 = Submit Answer | Tries 0/3 b. Now compute the 4 statistics below: Z X 2 F Z = X2 = F= Compute the values of the 4 test statistics. Submit Answer |Tries 0/3 Submit Answer Tries 0/3 Submit Answer | Tries 0/3 Submit Answer | Tries 0/3 Compute the p-values for each statistic. Assume the alternative for the Z and t test is 1-sided (Enter p-value as a decimal not a percent.): HA: slopepopulation > 0, and assume the alternative for the * and F is 2-sided: HA: slopepopulation # 0. p-value = X2 How many degrees of freedom? p-value = p-value = p-value = Submit Answer Tries 0/3 Compare the p-values for the Z and t. Why are they so different? Submit Answer Tries 0/3 O Because Z and t always yield very different p-values. Submit Answer Tries 0/3 Because n is only =5 and Z and t give quite different results for very small samples. Submit Answer Tries 0/4 c. Your co-worker also took a random sample from the same population and computed the 4 test statistics to test the null that slopepopulation=0. He gave you the 4 test statistics but forgot to tell you which was which. Here are the 4 test statistics in scrambled order: 9, 2.89, 3, 8.33 Which is which? Z= 1. x2 =[,t = , and F = Submit Answer | Tries 0/3