2. We will now have you try to determine the number of data points needed to guarantee a certain confidence on the proba
Posted: Thu Apr 28, 2022 7:32 am
2. We will now have you try to determine the number of data points needed to guarantee a certain confidence on the probability estimate of Problem 1. We slightly modify the notation of Problem 1 to make explicit the dependence of the estimate on N, i.e, let Pin = ΣΕ Ι: (Χ.) N Using the Chebyshev's inequality, determine an No, such that for all N > No, we have a guarantee that Pr( (ộin – p:)? <0.01p:(1 – p:)) >0.99, - Thus, we are required to find the size of the data required so as to ensure that the estimate is 0.01p(1 - p) close to the real probability with a likelihood of 99%.