Rik, Yes, I understand the logic in using the buy price as the divisor in a %gain calculation for a short position. I believe your method agrees with the IRS method also. Sounds like you've been thru this quite a bit in the past, so thanks for your patience.
Hopefully though, the optimization routine is using the "absolute" price move and not a "relative" price move to determine optimum indicator parameters. In other words, the optimizaion should be independent of the method used to compute %gain. Do you have any idea if that is the case? --- In [EMAIL PROTECTED], "Rik Rasmussen" <[EMAIL PROTECTED]> wrote: > The issue of calculating gains on shorts has been argued here in the past. > Originally, the program computed it differently. Several of us made the > argument to change it and the developer did change it. I believe it is > currently computed in the way we argued for. > > ================== > This message is from the archives: > > Message: 1138 From: Rik Rasmussen Received: Sun Jan 11, 2004 8:42 AM > Subject: shorts > > Profit on shorts calculation has been corrected in version 4.32. > > Back tests with long and short should give truer profit report. > > Rather than use entry and exit, now uses buy and sell prices. > > Rik Rasmussen > Best Charts Yahoo Group Moderator > http://groups.yahoo.com/group/bestcharts/ > > ======================== > > > > If you short a stock at $10.00 and later cover at $9.00 you bought at $9.00 > to cover and originally sold at $10.00. > > You invested $9.00 when you bought, sold at $10.00, made $1.00 profit. > > (sell price - buy price)/ buy price = profit > > ($10.00 - $9.00)/$9.00 = 11% profit > > I know some people will argue that shorting should be calculated > differently, but to me this is the only method that makes sense to me. > > The only difference, as far as I am concerned is the fact that I sell fisrt, > buy last. > > Rik Rasmussen > Best Charts Yahoo Group Moderator > http://groups.yahoo.com/group/bestcharts/ > > > > > -----Original Message----- > > From: redloonbass [mailto:[EMAIL PROTECTED] > > Sent: Monday, September 06, 2004 8:41 PM > > To: [EMAIL PROTECTED] > > Subject: [bestcharts] Backtest Error for Short Positions > > > > > > Wonderful program, however, it appears to me that the backtest > > routine incorrectly computes the % gain for short positions. The > > program appears to use the exit price as the divisor for short > > positions. > > > > The correct computation for %gain on a short position should be > > (Entry-Exit)/(Entry) and for a long position, (Exit-Entry)/ (Entry). > > In both cases the divisor should be the Entry price. > > > > Consequenty, for short positions, it is overestimating gains and > > underestimating losses. If the optimization routine is also using > > the erroneous computation, then it is not finding the true optimized > > parameters for the "short only" and "long and short" options. > > > > I am using version 4.33.2. Am I correct? > > ------------------------ Yahoo! Groups Sponsor --------------------~--> $9.95 domain names from Yahoo!. Register anything. http://us.click.yahoo.com/J8kdrA/y20IAA/yQLSAA/GHeqlB/TM --------------------------------------------------------------------~-> Yahoo! Groups Links <*> To visit your group on the web, go to: http://groups.yahoo.com/group/bestcharts/ <*> To unsubscribe from this group, send an email to: [EMAIL PROTECTED] <*> Your use of Yahoo! Groups is subject to: http://docs.yahoo.com/info/terms/
