a thousand runs on the particular back testing that I just ran rather than do them on all the trades. I ran back testing on same strategy with same parameters, and got different Monte Carlo results each time. They should be the same. Otherwise it defeats the purpose of running an MC to get average and median performance numbers. Would the new version of WLP due out at the end of this month do MC differently and correctly? Thanks.
Size:
Color:
In random environment that consists of a large number of randomized runs based on the portfolio simulation results, why do you believe that randomized results should be the same each time? You just do a thousand runs while there are millions of combinations of the bar by bar returns.
Size:
Color: