Assume all variable have expected value 0. 1. Suppose the observable variables x , y , and z are related by the linear model cov ( x : ) = cov ( 2 :) = 0 Suppose that & = g(2 ) for a known function g (-). Are the parameters & and a identified? Explain. 2. Consider the linear model y = BX +6, cov (x . 6 )=0 Suppose that y is measured with error, i. e. the data under study consist of x y, such that cov ( x,7) = cov(2, 7) = 0 What is the relationship between / and the regression coelficient in the linear model cov (x. 5)=0 Does your answer imply that measurement error is irrelevant to the empirical analyst, in this case