4 Logistic Regression with Interaction Term You are analyzing how the birth weight of a baby (normal weight=0, low weight=1) depends on the age of the mother (number of years over 23, e.g. a 25-year-old will have value 2 ) and the frequency of physician visits during the first trimester of pregnancy ( 0= not frequent, 1= frequent). You have also decided to include an interaction term for age and frequency. Your logistic regression coefficients are as follows: (a) (4 points) Discuss the meaning of each coefficient, and explain what does the coefficient of the interaction term show. (b) (3 points) Specify the logistic regression models when the mother visited the physician frequently and when they didn't. Explain how the mother's age affects the odds in each scenario. (c) (4 points) Compare how physician visits affect odds of low weight at ages 18, 23, 25, 28, 30, by calculating the odds ratio of low birth weight for mothers with frequent physician visits over those with non-frequent physician visits, in the following table (fill the "Odds Ratio" column in the table below). Note: for age, you should use number of years over 23. (d) (4 points) Interpret the numbers in the "Odds Ratio" column, considering the listed confidence intervals. Hint: what does an odds ratio of 1 mean (holding other predictors fixed)?. (e) (5 points) compare the "difference in probability" of low birth weight for mothers at ages 18, 23, 25, 28,30 , in the table below. Interpret your results and compare your interpretation to part (d)