Hey guys, I am using the "urca" and "vars" package to estimate a VECM. I already estimated the VECM with the ca.jo, put restrictions on the cointegration and loading matrix via "alrtest", "blrtest" and "ablrtest" and finally reesimated the restricted VECM with cajorls. Is it possible to transform this restricted VECM into a level's VAR version? As it can be done with "vec2var" and "ca.jo". Because I want to estimate in a second step the impulse response function in levels, rather than differences. (irf() or even SVAR()). Thanks a lot, dha! -- View this message in context: http://r.789695.n4.nabble.com/Restricted-VECM-VAR-representation-tp4638874.html Sent from the R help mailing list archive at Nabble.com.