On Mon, 28 Mar 2011, Grzegorz Konat wrote: > First of all, thank you [Jack] so much for the script - works > perfectly fine for me! In fact, the FEVD is available internally in gretl. I've now added an accessor, $fevd, so you can grab it as a matrix. > As far as the issue with different estimates is concerned [...] > > In gretl I do: > > vecm 2 1 X Y Z --ct > > However, when I use R's (urca): > > xyz.vecm <- ca.jo(gretldata[, c("X", "Y", "Z")], type = "trace", ecdet = > "trend", K = 2, spec = "transitory") /for all options in 'type' and > 'spec'/ > > and transform obtained VECM into VAR representation: > > vec2var.xyz <- vec2var(xyz.vecm, r = 1) > > the system I get has quite different parameters [...] The model you're estimating in R is equivalent to using the --crt flag (constant plus restricted trend) in gretl. If you use --crt you should find that the VAR coefficients agree pretty well. Allin Cottrell _______________________________________________ Gretl-users mailing list Gretl-users(a)lists.wfu.edu http://lists.wfu.edu/mailman/listinfo/gretl-users