Thursday, August 21, 2008

Short-Term Reversal Patterns Among Global Equity Indexes

A number of traders have commented to me on how choppy the market conditions have become. A strong movement seems under way, and then it just as strongly reverses.

As a way of looking simply at recent trading conditions, I went back to the start of 2007 and investigated three-day returns as a function of the prior three-day returns. Specifically, I looked at what happens when the market is up jointly on a one- and three-day basis (uptrending) and when it is down jointly on a one- and three-day basis (downtrending).

When the S&P 500 Index (SPY) has been up for the past one and three days, the next three days average a loss of -.30% (80 occasions up, 83 down). When SPY has been down for the past one and three days, the next three days average a gain of .22% (82 up, 51 down). If traders wait several days for a trend to assert itself and then jump on board, they are likely to start in the hole.

When we look internationally at the Europe, Australasia, and Far East (EAFE) stocks (EFA), when those are up on a one- and three-day basis, the next three days average a loss of -.27% (76 occasions up, 80 down). When EFA has been down over the last one and three days, the next three days have averaged a gain of .09% (74 up, 65 down).

Finally, when we examine emerging market stocks (EEM), we find that when they are up on a one- and three-day basis, the next three sessions average a loss of -.31% (86 up, 81 down). When EEM has been down over the last one and three days, the next three days have averaged a gain of .69% (79 up, 45 down).

Across the globe, short-term trend following has been hazardous for traders' wealth. Even longer-term traders need to take these reversal patterns into account, if only to size positions and set stops for expected heat.

RELATED POSTS:

Fading the Herd

Momentum and Reversal Effects
.

5 comments:

nico said...

I did a little further research on this, and YES, the market has an INCREASED tendency to reverse gains since early 2008.

Find the mini-study here:

http://is.gd/1NmV

Best regards & good trades!

Nico Richter

Woodshedder said...

Nico, I would have commented on your site, but I could find no way to do so.

I find that your conclusion that "selling after strength and buying after weakness didn't go anywhere during the last 5 years" to be false.

According to your entry criteria, what you publised is true. There are, however, many other entry filters for buying weakness.

Selling after strength and buying after weakness has worked well since the early 80s.

To determine this, you need to use a much more general entry criteria than you tested.

Paolo Pezzutti said...

Dr. Brett,
I find the concept quite interesting. Regardless the increase in volatility printed in the past months and a directionality mainly to the downside, runs both up and down have been quite short. The problem in this type of approach in my humble opinion is that the everchanging cycles could hurt you before you realize that something has changed. There is a need to have in place a monitoring system that tells you if the market is continuing to behave as expected or something is changing. In that case it gives you an alert. What is the length of data do you think should be taken into account both for validating a specific type of behavior and realizing that patterns of behavior are changing?
Paolo Pezzutti

nico said...

Woodshedder,

thanks for your comment, and I absolutely agree. I should have pointed out more clearly that I meant that contrarian trading AS TESTED there didn't go anywhere. I am fully aware of that 15-minute-study being a very small and incomplete look at contrarian trading in general.
I just tried to take a look at the question if the market has changed in that respect recently and if, maybe, contrarian trading has become "easier", as in that such an overly simple strategy might have begun to work better recently.

In no way did I mean to say "contrarian trading didn't work until recently". Oh my, if it hadn't, I would have almost no income :-). Oh, and sorry for there not being a commentary option in where I published the study. I just used evernote as a very quick outlet of this 15 minute approach towards looking at that question.

Thanks again for your thoughts; I like your blog btw

Woodshedder said...

Glad you enjoy the blog Nico.

I have to say that buying weakness has been good to me as well, hence my quick response in defense of the strategy!

I am working on a post right now that is piggybacking more on Dr. Steenbarger and other's recent work concerning reversals and trend. Stop by again if you get a chance. I hope you'll find it interesting.