Comparing results to Buy and Hold
Author: Creighton
Creation Date: 7/30/2014 12:57 PM
profile picture

Creighton

#1
Thanks for the help with my other question! I have a few more that I'd appreciate some help with.

I'm still learning the system and I don't really understand what the buy and hold results mean. And I don't understand how exactly to compare the buy and hold results to my trading strategy results.

I would have thought that buy and hold means all securities are bought at the beginning of the time frame and sold at the end. However, the buy and hold results are different when running the test using different strategies. I would think the "buy and hold" results should be the about the same for every test if the only variable that is changing is the test strategy since I didn't think the testing strategy had anything to do with the buy and hold results.

Also, when comparing strategies against each other, do I simply compare their profit per bar results? Or should I be more concerned with their individual ratios between profit per bar and buy and hold?

For example, I ran the default Dip Buyer strategy against the SP 100 for a 3 month time frame (4/29/2014 to 7/25/2014) using 60 minute bars the the resulting profit per bar was 4.8 compared to .93 for buy and hold. Using the same parameters with the EOM Trend Strategy, the profit per bar was .05 compared to .13 for buy and hold. I don't understand why the profit per bar for buy/hold is so different for the two strategies. The only variable I changed was the strategy which I didn't think had anything to do with buy and hold. Also, in order to judge effectiveness, should I be more concerned with comparing the profit per bar result between Dip Buyer and EOM Trend Strategy (4.8 compared to .05), or should I be more concerned with the respective ratios to their buy/hold results?

Sorry for the long question but I would really appreciate some help. Please let me know if my question is confusing in any way and I will try to clarify. Thanks.
profile picture

Eugene

#2
Let's make some guesswork:

1a - The difference is most likely in the position sizing.
2a - One of the strategies may have the Position Size preference saved (Preferences > Advanced Options).
2 - You could have unintentionally chosen to run one of the strategies in single symbol mode vs. portfolio backtest mode.
3a - Before opening the 2nd strategy, you could change a setting in the Preferences dialog (Commissions, Backtest settings, Slippage).
3b - Typed in a new Buy&Hold symbol there.

If the difference still remains, could you attach two full-size screenshots to prove it?
profile picture

Creighton

#3
Hm well I don't know what I was doing wrong yesterday, but I wasn't having the same problem today. I ensured that my settings matched your instructions and they did. For some reason today when I re-tested the strategies I got the same buy and hold results for every different type of strategy. This is what I had assumed was supposed to happen so I am glad the problem is fixed. Thanks for your help.

Also, I think I figured out the "profit per bar" part of my question. One thing I'm trying during testing is figuring out "profit per day" based on "profit per bar." This way I was hoping to be able to compare results across multiple time parameters. For daily bars, 1 bar = 1 day. For hourly bars, 7 bars = 1 day (7 hours in a trading day). So if one strategy (daily bars) resulted in $7 per bar, and another strategy (hourly bars) resulted in $1 per bar, then these strategies would have been exactly equal in terms of their profitability, correct? Do you see any problems with this that I may not be anticipating?

Thanks!
profile picture

Eugene

#4
QUOTE:
Hm well I don't know what I was doing wrong yesterday,

Most likely, selecting a different position sizing mode/option.

QUOTE:
Do you see any problems with this that I may not be anticipating?

Unless Robert (Cone) has a different opinion (which tends to be right), I'd say it's a roughly valid comparison. Something to keep in mind is that there happen short sessions and some less liquid stocks may not have a trade in minutes (maybe even hours).
This website uses cookies to improve your experience. We'll assume you're ok with that, but you can opt-out if you wish (Read more).