folks, is there a better way to convert a daily series of price returns to a monthly one than as follows: library(xts) a <- xts(rnorm(100) / 100, order.by = as.Date(1:100)) am <- diff(to.monthly(cumsum(a))[, 4]) # [, 4] for Close levels the problem with this is that the entire first month of returns is lost owing to the differencing. i would have to prefix the first month's returns manually, which seems a bit ugly. if i had the original series in levels, then there's no issue, of course, but all i have in this instance are returns series. cheers, murali _________________________________________________________________ Contest