[R] Fitting Gamma distribution

Marcosco01 marco.scortichini at bancamarche.it
Wed Jun 23 11:16:47 CEST 2010

I'm looking for goodness of fit tests for gamma distributions with large data 
sizes and for different data. 
I have a matrix with around 4.000 data values about losses and there is a
heavy right-tail in it. 
I have fitted a gamma distribution with "fitdistr". 

You can see the example: 


Errore in optim(x = c(5000, 5000, 5000, 5000, 5000, 5000, 5000, 5000, 
5000,  : non-finite finite-difference value [2] 

The problem is the optimization for the test with different data.

In summary 
- how to estimate the parameters? 
- is there another way to estimate the gamma parameters that doesnt depend
on the sample size? 

View this message in context: http://r.789695.n4.nabble.com/Fitting-Gamma-distribution-tp2265313p2265313.html
Sent from the R help mailing list archive at Nabble.com.

More information about the R-help mailing list