1. (10 points.) Trosiet chapter 11.4 Problem Set C, parts 1-3. Data: http://mtrosset . pages. iu. edu/StatInfeR/Data/cholesterol. dat Cholesterol Levels for Heavy Type A Men 293 201 312 250 246 197 268 224 23 23 254 276 234 181 248 252 202 218 212 325 Cholesterol Levels for Heavy Type B Men 3414 185 203 246 224 212 188 250 148 169 226 175 242 252 153 183 137 202 194 213 In the early 1960s, the Western Collaborative Group Study investigated the relation between behavior and risk of coronary heart disease in middle-aged men. Type A behavior is char- acterized by urgency, aggresion and ambition; Type B behavior is noncompetitive, more relaxed and less hurried. The following data are the cholesterol measurements of 20 heavy men of each behavior type. We consider whether or not they provide evidence that heavy Type A men have higher cholesterol levels than heavy Type B men. (i) Respond to (a) to (e) in Problem Set B. (a) What is the experimental unit? (b) From how many populations were the experimental units drawn? Identify the pop- ulation(s). How many units were drawn from each population? Is this a 1- or a 2-sample problem? (e) How many measurements were taken on each experimental unit? Identify them. (d) Define the parameter(s) of interest for this problem. For 1- sample problems, this should be p; for 2-sample problems, this should be A. (e) State appropriate null and alternative hypotheses. (ii) Does it seem reasonable to assume that the samples & and y, the observed value of XX, and Y,-.., Y, , were drawn from normal distributions? Why or why not? (iii) Assume that the X; and the Y, are normally distributed. (a) Test the null hypothesis derived above using Welchs approximate t-test. What is the significance probability? If we adopt a significance level of a= 0.05, should we reject the null hypothesis? (b) Construct a (2-sided) confidence interval for A with a confidence coefficient of ap- proximately 0.90