required number of simulation for monte-carlo analysis

My question is about the required number of simulations for Monte-Carlo analysis method, as far as I see for any allowed percentage error e.g.E=5, required number of simulation "n" is given by:

n= (100*zc*std(x)/(E*mean(x)))^2 (<http://www.dtic.mil/dtic/tr/fulltext/u2/a423541.pdf>)

where std(x) is standard deviation of resulted sampling, and zc is confidence level coefficient e.g. it is 1.96 for 95%. For eaxmple I run the simulation 100 times and then check resulted mean and std of 100 simulation actually represent accurate mean and std with %95 confidence level. When number of simualtion exceeds required n I terminare simulation otherwise increase number of simultion. But one time I run the simualtion 7500 times and create different mean and std of 100 sampling from 7500 simulation results (a moving average and std), even required number of simulation is always less than 100, but % error of mean and std compare to mean and std of entire results is not always less than 5%. % error of mean is less than 5% in most case but error of std goes up to 30%. What is the best way to determine number of required simulation without know actual mean and std for monte-carlo? also in my case subjected outcome of simulation is normal distributed, thanks in advance for any help

Answer 1

[锋 李]

so easy