STAT231 Study Guide - Quiz Guide: Confidence Interval, Exponential Distribution, Maximum Likelihood Estimation
Document Summary
For q4 ----part (iii), the alternative hypothesis is two-sided. Therefore there is a natural pairing of the data by factory. Factories with the best safety records before the safety program tend to have the best records after the safety program as well. The analysis of the data must take this pairing into account. Therefore the model assumed for analyzing these data is: Yi = + ri,whereri ~ g(0, ), i =1,2,12 where and are unknown parameters. (cid:1)(cid:3)(cid:2) (cid:1)(cid:3)(cid:3)(cid:2) (cid:1)(cid:3)(cid:3)(cid:3)(cid:2) Explain what the parameters and represent. Test the hypothesis of no difference due to the safety program, that is, test the hypothesis h0 : =0 : (1+2+2+1+2+2= 10 points)survival times for patients treated for a certain disease have an exponential distribution with mean . Under the standard treatment the mean survival time is 37. Ten patients receiving a new treatment survived for the following times (in months):