We consider the problem of estimating the first k coefficients in a regression equation with k + 1 variables. For this problem with known variance of innovations, the neutral Laplace weighted-average ...