Dear all, I'm using the function lm.gls but I'm puzzled by the very limited output. I do get the regression coefficients but not their standard errors. Neither summary(), anova() or vcov() works properly for a lm.gls object. Preferably I look for a function like vcov that will give me the variance-covariance matrix for the estimated coefficients, but just the standard errors will do. I'm aware of the gls(nlme), but I use a correlation and weight structure that is not easily implemented in this function. Thank you in advance Ida -- View this message in context: http://r.789695.n4.nabble.com/vcov-for-lm-gls-tp4185348p4185348.html Sent from the R help mailing list archive at Nabble.com.