Dear Gretl users, I have estimated my system of equations and taken the matrices of the residual sum of square (from 2 steps). Now, I should make an LM test as follow: LM=T(p-tr(RSS1^-1 RRS2)) Notices that RSS1 and RSS2 are 4x4 matrices. 1) Question: Is it correct if I write a code in gretl as: LM=T(p-tr(RRS2/RRS1)) ? 2 Question: since I never used to write a code to test but I usually test a model by a click, then a question rises. Is it correct that the output of the test is a series with the same value? 2) I suppose that I can obtain also the p-value of the test. Can somebody suggest me the code? Thank a lotValentina From: valentina81c(a)hotmail.it To: gretl-users(a)lists.wfu.edu Subject: system of equations Date: Thu, 10 Oct 2013 19:11:17 +0000 Thanks. However, the command "series res= $uhat" works only whether I estimate my system only equation by equation. To be clear, it works whether I estimate 4 different equations by ols and then collect step by step the residual of each regression by the above command and the RSS. It may work, but then I need the matrix residual sum of squares of the system of equations to pass at 2nd step to compute. Because at end I have to compute the test statistic LM=T(p-tr{RRS0^(-1) RRS1} Has somebody any suggestion to obtain in a simple way the matrix residual sum of squares?Thanks

Am 10.10.2013 21:11, schrieb valentina colombo: I don't think I really understand what you want to do in the end, but anyway some remarks that might clarify some things: - yes you can get the residuals from a system via $uhat, as explained in ch. 28 of the guide. You get a matrix type variable, not a series. You also get the estimated cross-equation covariance via $sigma, which may also help. - Are you perhaps trying to do the test which is available directly in $diagtest along with $diagpval (again see ch. 28)? -sven