... very large.Copyrightc 1999, Russell Davidson and James G. MacKinnon 18 Regression ModelsAs a consequence, much of the art of constructing, or specifying, a regressionmodel is deciding which ... small.Copyrightc 1999, Russell Davidson and James G. MacKinnon 24 Regression ModelsNow suppose we take logarithms of both sides of (1.25). The result islog yt= β1+ β2log Xt2+ β3log Xt3+ ... among others, Billingsley (1979).Copyrightc 1999, Russell Davidson and James G. MacKinnon 16 Regression ModelsConditional ExpectationsWhenever we can describe the distribution of a random...