December 16, 2013

Stop Losses and Equity Curves

This extended research is from two readers’ requests. Request one is adding stop losses on the “Monthly S&P500 Stock Rotation Strategy.” Request two is seeing the equity curves from “Percent S&P500 Stocks Trading Above MA50 as Market Timing Indicator.”

Stop Losses

Continuing from the post, we are adding a maximum stop loss. The stop is evaluated at the close each day with the exit happening at the close.  The tested stops are 5%, 10% and 15%.

Worst N-Month Ranking

ReaderReqs131216A

Using the Worst N month ranking, we can see that the stops help either by raising CAR and lowering MDD. Normally close stops tend to be the worst but the 5% stop tends to be the best. Normally stops hurt mean reversion strategies. The longer hold periods may be the reason they help in this case.

Best N-Month Ranking

ReaderReqs131216B

For the best N-month ranking, stops appear to hurt the all results.

 

Equity Curves

ReaderReqs131216C

From “Percent S&P500 Stocks Trading Above MA50 as Market Timing Indicator,” Ronen asked to see equity curves for the following variations: Buy & Hold, Below 20, Below 40 and Below 80. One can see how they all get hit in the 2008 Crisis. Below 40 was the first to get to new highs and has continued to perform well.

Spreadsheets

If you’re interested in a spreadsheet of my testing results, enter your information below, and I will send you a link to the spreadsheet. Included are yearly breakdown of returns, 3, 6, 9, & 12 month ranking, and 10, 20, & 50 positions.

Final Comments

This will be the last post of the year. I will busy with FTC Robotics with my son and visiting family. To be notified of future posts, sign up for email alerts. Keep the great comments and research suggestions coming. I enjoy doing the reader request research. Enjoy your holidays!

Backtesting platform used: AmiBroker. Data provider:Norgate Data (referral link)

 

 

 

 

Click Here to Leave a Comment Below

Darren - December 17, 2013 Reply

It would be interesting to see the results with Van Tharp style ATR volatility and “Maximum Adverse Excursion (MAE) over last X days” based stops. Does the stop become beneficical (reduce max drawdown) for mean reversion when the stop is placed k x MAE away from entry price?

Rick - December 23, 2013 Reply

Great work Cezar – thanks

It would appear you have the Best and worst N-month ranking table headings (above) the wrong way round. As the results (based on CAR and MaxDD) with 0 stop loss correspond to the opposites in your previous post about monthly SP500 rotation strategy.

This means that applying a stop to the worst N-month ranking would be beneficial in some cases, however, applying it to the best N-month ranking would be counter productive in all instances.

    Cesar Alvarez - December 23, 2013 Reply

    You are correct. It has been corrected on Dec 23, 2013. Thank you for catching that.

Rick - December 23, 2013 Reply

Cesar,

Glad to be of help.

Your results are very thought provoking and interesting for someone who does not have the skill/resources to run such tests.

I would be interested to see the results of a couple of further tests if you have the time and your database is able to test them.

1. Academic studies suggest that the momentum effect (Best N-month ranking) is strongest in small cap stocks and weakest in large cap. Therefore if you were to repeat these studies on say the S&P400 MidCap stocks and then the S&P600 SmallCap stocks (in theory the momentum effect should be stronger) they should show an even stronger return to MidCap and SmallCap for the Best N-month ranking. But a weaker return to the Worst N-month ranking (because negative momentum will also be stronger) i.e. mean revision should get weaker… at least that is the theory!

2. Since stop losses only hinder the Best N-month ranking, then why use them? However most “expert” traders say use them. My viewpoint would be that they are only useful for longer holding periods (since your monthly re-balancing effectively cuts losers short anyway). If you were to re-balance the portfolio say every 2, 3, or 6-months, then it is possible that you might see some benefit for using stops (just a guess) whilst reducing stock turnover. What effect this would have on overall return when compared against monthly rotation would be interesting.

best wishes, Rick.

PV - September 9, 2014 Reply

For the timer, what about limiting the min% above MA to say 15?
So still do below 80,40 and 20 but aswell as that there still must be 15% above their MA50.
That might be a nice ‘move to cash’ addition.

Marco - September 25, 2014 Reply

A very good example of the so difficult task in stop loss sizing.
http://nightlypatterns.wordpress.com

Leave a Reply: