cameronmoss3
09-18-2002, 05:28 AM
Hi I was wondering if anyone here can help. I was re-reading through the Mahler note on Chi-Square statistic test when i came across this:
In an example about grouped data from a Burr distribution, he
mentioned that the number of claims observed in a particular interval WILL be approximately Poisson. He did first mentioned that the mean number of claims for the interval [a,b] is N*{F(b) - F(a)}, where F(x) is the assumed distribution function and N is the total number of claims across all intervals.
Now, why is this so? Is this unique for the Burr Distribution (ie the frequency of claims in an interval is approximately Poisson?)
Thanks in advance. :-? :-?
In an example about grouped data from a Burr distribution, he
mentioned that the number of claims observed in a particular interval WILL be approximately Poisson. He did first mentioned that the mean number of claims for the interval [a,b] is N*{F(b) - F(a)}, where F(x) is the assumed distribution function and N is the total number of claims across all intervals.
Now, why is this so? Is this unique for the Burr Distribution (ie the frequency of claims in an interval is approximately Poisson?)
Thanks in advance. :-? :-?