The Motley Fool Discussion Boards
Investing/Strategies / Mechanical Investing
|Subject: Re: The Best Measure for the Best Blend||Date: 11/24/2007 7:39 AM|
|Author: Zeelotes||Number: 204271 of 265627|
Maybe it is just my own denseness, but I'm not clear what the columns represent in your data. Your daily/weekly etc. columns: is this the granularity of all the calculations, or the one on which you are optimizing, or the resulting measure after optimization?
Right now almost everyone on this board has one option for optimizing blends and that is using Jamie's backtester which outputs monthly granularity. Of course, I have daily data, so I can calculate off of daily, weekly, monthly and yearly, or anything else for that matter. In the table you are seeing above, I've shown the differences in the results from using these various granularity levels.
My question really is this: if there are significant differences in the result of the optimization, using daily Sharpe as the measure, regardless of whether you are optimizing on Sharpe, GSD, or anything else: then might not there also be differences depending on whether you rank the optimization by final Sharpe based on annual instead of daily data?
Yes, there most certainly are differences as the tables above show. It can make a huge difference. Of course, the problem is that the vast majority do not have any other option besides monthly or yearly data. They can't get ahold of weekly or daily, so in some measure, although I find daily to be optimal, that doesn't help too many. Of course, if you can get the daily data from Robbie's backtester, than I'd guess that would resolve this problem.
Are you clear now, or only as clear as mud?
|Copyright 1996-2017 trademark and the "Fool" logo is a trademark of The Motley Fool, Inc. Contact Us|