R help - Feb 2012 - error in fitdistr

Hi dear,

I want to estimate d.f for Chi-squared distribution:

est.chi[i,]<-c(
fitdistr(as.numeric(data2[,i]),"chi-squared",start=list(df=1))$estimate)Warning
message:In optim(x = c(7.86755, 7.50852, 7.86342, 7.70589, 7.70153,
7.58272,  :
  one-diml optimization by Nelder-Mead is unreliable:
use "Brent" or optimize() directly


Who can help me to solve this problem?

Best whishes,
Soheila

	[[alternative HTML version deleted]]

On Feb 23, 2012 at 11:19am Soheila wrote:
> Who can help me to solve this problem?
> est.chi[i,]<-c(
>
fitdistr(as.numeric(data2[,i]),"chi-squared",start=list(df=1))$estimate)
> Warning message:In optim(x = c(7.86755, 7.50852, 7.86342, 7.70589,
> 7.70153, 7.58272,  :  one-diml optimization by 
> Nelder-Mead is unreliable: use "Brent" or optimize() directly
The warning message tells you to use "Brent" rather than the default
Nelder-Mead. So do that.

##
?optim
est.chi[i,]<-c( fitdistr(as.numeric(data2[,i]),
densfun="chi-squared",
start=list(df=1), method="Brent")$estimate)

Regards, Mark.

-----
Mark Difford (Ph.D.)
Research Associate
Botany Department
Nelson Mandela Metropolitan University
Port Elizabeth, South Africa
--
View this message in context:
http://r.789695.n4.nabble.com/error-in-fitdistr-tp4413293p4413459.html
Sent from the R help mailing list archive at Nabble.com.