[R] pel fitdistr how to estimate distribution parameters?

victoryday v.deringer at gmx.de
Thu Feb 27 10:33:55 CET 2014

I`m using fitdistr() to estimate parameters for different distributions of my
data set.

Now I have problems to estimate the weibull Distribution and others. I found
the recommendation to use the pel…() function.

My question is: If I estimate parameters with the pel…() function. How can I
use them to create a theoretical distribution?

Like for normal distribution, exponential distribution, gamma distribution
and so on.
For example: I got this parameters:

> pelnor(x)
         mu      sigma
 0.09968345 0.12117581

> pelexp(x)
          xi       alpha
 -0.03704881  0.13673225

x ist my data set

I try to estimate parameters for different distribution based on my real
data set and then to create a theoretical data set based on the estimated
parameters. So I can compare to which distribution my data set match at
most. For this I use the chi quadrat test.

View this message in context: http://r.789695.n4.nabble.com/pel-fitdistr-how-to-estimate-distribution-parameters-tp4685919.html
Sent from the R help mailing list archive at Nabble.com.

More information about the R-help mailing list