Re: [R] Errors with maxNR: NA in gradient

Hello, Try using an equal sign after 'sigma': algo=maxNR(loglikelihood, start=c(sigma=8.686603,beta1=-4.976215,beta2=7.313875)) Hope this helps, Rui Barradas   Citando Emma MONTARSOLO : > Hi ! > > I use maxNR function (from maxLik package) to estimate parameters > via maximum likelihood a

[R] Errors with maxNR: NA in gradient

Hi ! I use maxNR function (from maxLik package) to estimate parameters via maximum likelihood and Newton–Raphson method. Context : I have a variable Y which follows a Gumbel distribution of parameters Beta1X1 + Beta2X2 (location parameter) and sigma (scale parameter). I have to estimate sigma,