To estimate Copula using canonical maximum likelihood method we need first tranforming data to uniform variates. I found this code written by Jun Yan: cbind((rank(dat[, 1]) - 0.5)/n, (rank(dat[, 2]) - 0.5)/n) where dat represent the considered data. Why do not use this formula rank(dat)/(n+1) is it the same? -- View this message in context: http://www.nabble.com/Transform-data-to-uniform-variates-tp17985481p17985481.html Sent from the R help mailing list archive at Nabble.com.