Generalized Pareto Distribution: generate theoretical data given estimated parameters

[Quantopic]

Hi all,

I estimated the scale and the shape parameters using empirical data on the base of the PROC SEVERITY.

Now, by using the estimated parameters XI and THETA for the generalized Pareto distribution, I want to compare the empirical distribution with the theoretical one.

First of all, I would like to generate data extrapolated by the GPD distribution on the base of the estimated parameters and plotting the theoretical data against the empirical one.

Could someone suggest a simple way to do that?

Thanks in advance for the help.

[DarthPathos]

Hi and welcome to the community!

I've started playing around with simulating data and found the following to be really helpful:

http://support.sas.com/resources/papers/proceedings15/SAS1387-2015.pdf

http://www.statsblogs.com/2014/06/04/simulate-lognormal-data-with-specified-mean-and-variance/

Also, Rick Wicklin has written a book on simulating data in SAS that I found very good. 

Good luck and hopefully these help!

Chris

[Quantopic]

Thanks for replying, but I need exactly for an hint to simulate GPD data.

In the papers you posted above, you can simulate othe kind of distribution but there is no suggestion about to simulate GPD distributed data.

Thanks anyway.

[DarthPathos · Accepted solution]

My mistake, I (incorrectly) assumed that a GPD was something else.  

Trying again:

https://communities.sas.com/t5/General-SAS-Programming/Rand/m-p/298418 - Another user asked the same question and I think got the answer they needed.

Sim Data Book - Google Books with the page from the book I mentioned where the GPD is used.  

Hopefully these help and good luck 🙂

Chris

[Quantopic]

Solved!! 

Thanks for the help!

[DarthPathos]

Anytime, it's why we're here as a community!

Happy coding!

Chris