I would like to know what forum members think about the value of optimizing strategies. For example, trying different numbers for price sales ratio, EPS growth percentage, ROE etc. in any strategy until we get better risk adjusted returns. I know some may consider optimization as nothing but curve-fitting. Is anyone using optimization of their strategies in real life and getting better returns as a result? It would also be nice to mechanize the optimization process in the software wherein the system tries different numbers out of a range of numbers specified by the user for different parameters and recommends the best system based on different outcomes for a given time period. Thanks.


Attach file (image (.gif, or .jpg), screen (.und) or report (.rpd))
ilikemyodds
at 11/22/11 10:11 PM : RE: Optimization

Thomas
at 07/07/12 3:21 PM : RE: Optimization

Thomas
at 11/23/11 6:10 PM : RE: Optimization

Thomas
at 11/24/11 10:36 PM : RE: Optimization

ilikemyodds
at 11/24/11 9:23 PM : RE: Optimization

Thomas
at 11/24/11 5:45 PM : RE: Optimization

ilikemyodds
at 11/24/11 3:36 PM : RE: Optimization


Thank you very much for your detailed reply. I must confess that some parts of your reply are difficult for me to undestand (I am not an investment professional). Nevertheless your conclusions resonate with me. Mark Hulbert, the publisher of Hulbert Financial Digest has made a remark in one of his recent articles that it is nearly impossible to make more that 15% per year over the long term from investments unless one includes hedging techniques. He has reached this conclusion by following the recommendations of some of the best financial advisory letters over long term. Even Warren Buffet's track record is only 20% per annum on an average. Your back testing also seem to quote 15% annualized return over 32 years. I agree that 12+ years is not long enough to test the robustness of a strategy. James P. O'Shaughnessy in the latest edition of his book What works on Wall Street recommends testing a minimum of 25 years of data before accepting that a particular strategy will be good long-term. Finally as you say that if the data you are testing is not 100% reliable and is not free from the usual biases, then the conclusions you draw from testing such data are unreliable and could cost you money.

By the way what is the name of your back testing program and is it available commercially? What would you recommend as the best reliable financial database to use for backtesting purposes? Thanks.


Attach file (image (.gif, or .jpg), screen (.und) or report (.rpd))

For questions or more information or to Order Today, call 1-800-767-3771, ext. 9392 or 1-312-630-9890, ext. 9392
You can also e-mail our Sales Support at: kevinm@zacks.com
Copyright 2012
Zacks Investment Research
111 N. Canal St., Suite 1101, Chicago, IL 60606
(800) 767-3771 ext. 9392