random 100k rows of kernel numbers, each row has Poisson number of random numbers

Hi,

I am trying to do something and I need help.

So, this is for insurance analysis. I have x number of accidents that may result in a claim (modelled as Poisson distribution). And I have a Y kernel distribution of money value of the accident.

When I multiply X*Y I am assuming that all accidents in one year cost Y, which gives me problems computing the due claim (ex.:if that average is bellow mininum then no claims result for that year, although in reality at least one accident should be insured).

So, I figured I should create a matrix of size (number of samples)*(maximum number of accidents in year) that would give me for each row/sample, p random kernel numbers, where p is the first value of the Poisson distribution.

Something like this: kernel: [random random random n/a n/a random n/a n/a n/a n/a random random random random n/a ...] Poisson: [3 1 4 ...]

Can I do this??

Thanks a lot,