Posted By: Roy   Date: Thu Dec 11, 2008

Title: Just wondering

Message:
  It seems the parameters change everyday on the new 21.1 model. Is that good or bad. I was looking back on the past results of our present 20.1 model and said to myself, boy, this is really good. A 66% profit in '07 with four small losses totaling 2.25%. A 70% profit in '08 with four small losses totaling 2.89% before the only trade we had the bad fortune to participate in, the trade of 11/11 for a loss of 4.73% It seems the 20.1 has a hard time trading in volitile markets having missed the down turn at 9/30 and the last months' moves. Could it be that if the market ever returned to normalcy it would do well? For now I'm wondering why after all the hard work proving a new program, when it's finally put out it doesn't perform like we would expect it to. You higher math guys might have the answer to that.
  Roy

Response:
  Development work on version 21.1 is complete and now the optimization program is trying to find the best set of trading parameters. Currently, there are two competing sets. One set stops out more frequently than the other but has a higher return. The drawdown for both sets is about the same. Over the past few days, you have seen some preliminary results with these sets. I now plan to go with the fewer stops set but may reconsider this before the changeover on Saturday.

 As you have noted, this has been a difficult market to predict. I'm not sure that I would expect a return to "normalcy" until the credit problem is resolved and we have some expectation of higher earnings. Speaking of expectations, I would expect any model to make mistakes. That is why it is so important to set stops. Our models use stops as part of the overall trading strategy.