Dear R gurus, Could somebody enlighten me on how to estimate the parameters of ARCH process? I tried both garchFit and garch functions, but no one works for me. R version 2.6.1, Windows library(fGarch) set.seed(5) # simulate ARCH(3) rr=garchSim(model = list(alpha = c(0.1,0.2,0.6)),n=400) rrm=rr-mean(rr) fit = garchFit(~garch(3, 0), data = rrm) # ends in Error in solve.default(fit$hessian) : # Lapack routine dgesv: system is exactly singular library(tseries) summary(garch(rrm, order = c(0,3))) # never ends In fact, the example given in the garch function Examples section does not work too. Many thanks, Remi [[alternative HTML version deleted]]