Veteran
Posts: 183
Location: USA: MD, Columbia | Hello all,
I wanted to share below as found interesting findings from what I learned. I also wonder if this could be possible why Nirvana may not be able to provide an appropriate solution to position sizing algo. The article also coincides with Portfolio Simulation mentioned in Jim's video's.
I have attached the excel spreadsheet mentioned, in event, it was taken off the site.
http://www.iitm.com/Buying-Power-Spreadsheet-Tool.xlsx (for the excel file)
http://www.iitm.com/Weekly_update/Weekly_534_July_ 13_2011.htm
back issues - http://www.iitm.com/weekly_update_backissues.htm
excerpted ...
"What was your paper about?
I had been doing a lot of back testing and found some pretty good looking systems. However, when I added in position sizing™ rules to the testing, I was stunned to see how severely the performance deteriorated. It was so bad that at first I thought I had mathematical errors in my spreadsheet. After looking into it I saw that the results were bad because the system had to skip lots of trades. Missing trades really hurt the profitability of the system.
As I looked into the root cause of why so many trades were not taken, I saw there was a limit to the buying power of my capital—sometimes there was not enough money to take all the trades because the signals were clustered together. "
Attachments Buying-Power-Spreadsheet-Tool.zip (396KB - 15 downloads)
|
Owner/Admin
Posts: 3925
Location: USA: GA, Lawrenceville | Thanks, Salim!
I too am a supporter of Van Tharp's focus on position sizing. His newsletter is free, and I find something truly useful in about every second or third issue. IITM.com
I have used his ideas and folded in a lot if my own, re approaches to risk control and money management. Some of it can be implemented in OT, right now - and I have been assured that in the not TOO distant future (hopefully sometime next year) more will be possible along these lines in OT.
I will spend considerable time on this in the Video Training about OmniScript Utilization, starting up in about 3 weeks (after the Optimizing OT mini-series is done). There are some nifty ways to incorporate both equity limits and volatility based risk factors into OScript-calculable position sizes. And, from there, do some truly useful liquidity analysis and filtering.
The spreadsheet tells the story well. PortSim does a lot of that kind of work now, and Ed has promised PS will "blossom" into a much more robust money-management tool. In the meantime, we'll make the most of what we can do now.
Again, thanks for your post! |