Howard, I want to learn a little about significance testing for non-normal distributions ... where should I look i.e. what N-N method is most relative to trading distributions?
(I think you recommended some before but I am sorry to say I didn't make notes). Thanks. --- In [email protected], Howard B <howardba...@...> wrote: > > Hi Brian, and all -- > > You wrote: > "There are at least five points in Keiths post. > > The one that I recall you making previously is in agreement with Keith when > he said: > > >snip< If I have a system with good OOS performance, why should I care what > the IS performance is? >snip< > > AFAIK you haven't commented, in this forum or your QTS, on the other points. > > I will assume that the above point is the one you are re-endorsing." > > > ---------------------------------------- > > > Yes, that was the point I was re-endorsing. > > > I am not planning to comment on all the comments. > > > If there are is something that someone specifically wants me to comment on, > please write a _short_ posting asking the question very _clearly_, and make > it clear that the question is to _me_. > > Thanks, > Howard >
