here's the question

4. We consider an altered regressiOn problem for random variables X and Y. Let 2:1, . . . , :22", be the values observed from a modem sample of X. We aSSume that the conditional expected value of Y depends quadradioelly (rather than linearly) on the value of X, and that the conditional expected value of Y is 0 when X is 0. In other words, we assume Y; = ag-E + 5i: where is unknown, 81, . . . ,5\" are independent and identically distributed N (0, 0'2), where a2 is unknown. 8. . Find the maximum likelihood estimators for )9 and 02. Note. You must show your work deriving the formulas of the MLE's, but you may skip the details involving the second derivative of the likelihood function. (I) (II) b . We apply our model to the two sets of data depicted above, both of which have the same regression curve y = 2:2. Determine which data set has the larger 02. Give a convincing but succinct argument to justify your