Posted by: timetraveller May 5, 2007
Making Money with Eminitrader
Login in to Rate this Post:     0       ?        
Bri, just sent you an email. Read it. Also, I want to do a bit of quantitative research over the summer if I have the time and resources. The idea is still vague but it has to do with the return of the S&P 500 or some "hypothetically" named index that contains the various stocks from 1960 till present and see how the returns would have varied had other stocks been selected in random for such an Index. So I'll be using something called "Monte Carlo" simulation. I would need access to a database of stocks with the following: 1. A database of midcap and large cap stocks since 1960.Midcap because I want to make it more generalized. S&P 500 is all large cap) 2. A database that has their daily open and closing values. 3. Also, the ability to alocate them into various sectors and industries. Factors to be considered: -The S&P 500 initially had market value weighting. So till 2005 i'll need to use that method but after that, i'll need to use the "shares available for trading" method for those companies. -I'm pretty sure the market value weighting would not be much of a problem but having access to the "shares available for trading" criteria might be difficult since there may only exist a database that contains such figure for the current S&P 500 stocks. For others that weren't included, it may not exist. - I will also need to randomly throw discard a few stocks at a time to add NEWER stocks and hold them for a specified period of time, all done randomly. If anybody has attempted this kind of simulation, please let me know. also, if there's a quantiatively trained person here, especially in mathematics with knowledge of coding in MATLAB, that would be awesome!!! This s tougher than it even seems. For a simulation run at least a 1,000,000 times, it would probably eat up my computer for 2-5 days since there will be over a 1,000,000 different returns computer for each trial of randomly selected stocks and their returns till present. HELP me out. I don;t think I can do this on my own.:D:D the reason I want to do it is because I wanna see whether there would be ay significant difference between randomly choosing stocks and doing a thorough analysis. Now I respect such research, i truly do. But I want to do this for purely quantitative reasons and for fun. Have a good evening!!!!!
Read Full Discussion Thread for this article