Statistical uncertainties Monte Carlo

[Lian1985]

Hi!

I'm newbie in Monte Carlo. My question concerns the statistical uncertainties of my simulations.

Let's say that the result of my Monte Carlo is a certain distribution A (e.g. number of particles as a function of the depth in a target) and I have run NA events to generate that distribution. Then, I run the same code with NB events and I obtain the distribution B.
Can I calculate the statistical uncertainties of my results based on only on these information?

Thank you for your help!

 

[mathman]

You can get an estimate of statistical uncertainty, although there will be question of how good it is. It will depend partly on how large NA and NB are - the larger the better.

Suggestion: move to the statistics forum.

 

[Khashishi]

At the very least, you will be limited by shot noise due to quantized number of particles. Typically, this gives you a Poisson distributed noise, with error which scales as sqrt(N), with N the number of particles. Details will depend on what your code is doing. If your distribution is represented as a number of particles per bin, then the error is at least sqrt(number of particles in that bin) unless some averaging was done in the code.