Question 1 (10 points): Write down the logistic regression equation for the Natural Logarithm of the odds (of survival) Question 2 (10 points): What is the P-value for each of the partial regression coefcients in the variable equations table, and are these coefcients signicant at alpha = 0.01? Question 3 (20 points): Below are two observations (cases) taken from the data sample, with the values for the Weight, Humerus and Length (explanatory variables) and for the STATUS (binary response): Weight Humerus Length STATUS g I survived 0 perished) Case 1 24.5 0.687 154 I :(survived) Case 2 31.0 0.765 165 0 :(perished) The above are the observed cases. For each observed case, compute The predicted probability (P) of survival, write down the equation, how you did it. The predicted outcome (1 or 0), when the cutoff value is 0.5, write it down, how you did it. Is the prediction correct in case 1? Is the prediction correct in case 2? Question 4 (10 points) The classification table shows that the correct overall classification rate from applying this logistic regression model to the sample data is 68.4%. l ; How is this value computed from the numbers in the table, and what does it mean? The SPSS logistic regression results for the Bumpus data are the following: Classification Tablea Predicted STATUS1 Percentage Observed 0 1 Correct Step 1 STATUS 0 39 25 60.9 1 18 54 75.0 Overall Percentage 68.4 a. The cut value is .500 Variables in the Equation B S.E. Wald df Si . Ex B Step 1*1 WEIGHT -.553 .179 9.559 1 .002 .575 HUMERUS 59.782 12.631 22.402 1 .000 9186433651320 8790000000000 .000 LENGTH -.185 .053 12.158 1 .000 .831 a. Variable{s} entered on step 1: WEIGHT, HUMERUS, LENGTH. 4 a l l The binary response variable is the STATUS, with a value = l (the bird survived), and 0 (bird perished). Weight, Humerus and Length are the explanatory variables. Note that the results in the variables table assume that the intercept is zero, so the binary logistic regression equation has no intercept. Question 1 (10 points): Write down the logistic regression equation for the Natural Logarithln of the odds (of survival)