We consider the problem of identifying the political ideology of a politician from their voting record. the data consists of the votes of members of the u. s. house of representatives on 16 different issues in 1984. we are interested in predicting whether politicians are democrats or republicans based on their voting record. we will use 425 politicians to train the model, and 10 politicians to test it. we model political affiliation as a random variable a, and the votes on the 16 issues as the random variable v1, v2, . . , v16 (assume the vote outcome is binary: yes or no). a. if you make no conditional independence assumptions, how many parameters do you need to represent the joint pmf of vi, v2, v16; and a? why is this problematic? b. if you assume that the votes are conditionally independent given the affiliation, what is the minimum number of parameters that you need to represent the joint pmf? justify the assumption intuitively. c. express the conditional pmf of a given vi, v2, vi6 in terms of the parametrization of the joint pmf that you chose in the previous question.