Hello,I am using [R] to calibrate some vector autoregressive models (using the mAr package). I know there is a test comparable to the ljung-box or portmanteau test to test for the normality of residuals, but I cant find the precise formula on the web and of course I can't find a package dealing with that. Could so please help me out with that?also, I want to calibrate my VAR on a "discontinuous" time series: my series are daily but the process has month intervals - so I would like to calibrate using roughly 20 different "monthly series" obtained by stripping the daily series - my objective function would be the sum of 20 palin objective functions. Does that make sense?Thanks for any help!