R help - Jan 2013 - Assistant

Good-day Sir,

I am R.Language users but am try to? estimate parameter of beta distribution
particular dataset but give this error, which is not clear to me: (Initial value
in "vmmin" is not finite)
 beta.fit <- fitdistr(data,densfun=dbeta,shape1=value , shape2=value)
 kindly assist.
 expecting your reply:

You're not giving people much to work with. I googled the error, and it
seems to come from the call to optim and has likely to do with bad starting
parameters.

That said, the documentation of fitdistr doesn't suggest it even supports
"dbeta", there is only a "beta" mentioned.

On 22.01.2013, at 17:07, Adelabu Ahmmed wrote:
> Good-day Sir,
> 
> I am R.Language users but am try to  estimate parameter of beta
distribution particular dataset but give this error, which is not clear to me:
(Initial value in "vmmin" is not finite)
> beta.fit <- fitdistr(data,densfun=dbeta,shape1=value , shape2=value)
> kindly assist.
> expecting your reply:
> 
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.

Hello,

You are calling the function in a wrong way. In the case of a beta fit, 
densfun should be the quoted string "beta" and the initial parameter 
values are elements of a named list. Like this:


library(MASS)

x <- rbeta(1000, shape1 = 2, shape2 = 0.5)
fitdistr(x, densfun = "beta", start = list(shape1 = 1, shape2 = 1))


As for your error, I only got it if the data clearly can not fit a beta.

y <- rgamma(1000, shape = 2, rate = 0.5)
fitdistr(y, densfun = "beta", start = list(shape1 = 1, shape2 = 1))
Error in optim(x = c(6.19809706003757, 2.32632108817696, 
3.60844436009277,  :
   initial value in 'vmmin' is not finite


So revise the way you call fitdistr and then, if the error persists, 
revise the parametric distribution to be fitted.


Hope this helps,

Rui Barradas

Em 22-01-2013 16:07, Adelabu Ahmmed escreveu:
> Good-day Sir,
>
> I am R.Language users but am try to  estimate parameter of beta
distribution particular dataset but give this error, which is not clear to me:
(Initial value in "vmmin" is not finite)
>   beta.fit <- fitdistr(data,densfun=dbeta,shape1=value , shape2=value)
>   kindly assist.
>   expecting your reply:
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>