Edit: daily signals for this strategy are available in the State of the Market report.
This week I’m focusing on the TED spread and what it can tell us about the stock market. Not familiar with the TED spread? Read more.
In my last post, I looked at the TED spread as a lagging/concurrent indicator and showed that it’s been inconsistent at best. After that not-so-glowing review, I didn’t have high hopes for the TED as a leading/predictive indicator. But in this report I’ll show that, interpreted correctly, the TED has done a pretty good job at predicting monthly moves in the stock market.
[logarithmically-scaled]
The blue line above shows the results of trading the S&P 500 (frictionless) using the following strategy from 1988 to date: go long the S&P 500 following any month when the 6-month exponential moving average (EMA) of the TED rose. The red line shows the inverse strategy (long when the EMA fell).
And for the number-lovers:
This is a contrarian strategy. Put simply, when recent TED readings are higher (indicating higher perceived interbank credit risk), we are buying the stock market. This is counterintuitive (i.e. we’re buying when things are getting worse), but as we’ve shown before, the stock market is usually darkest just before dawn.
Two additional comments:
First, note that October bucked the trend. The TED spiked very high in September, but the market most definitely did not rebound in October. We’ve seen this in just about all contrarian indicators, but over a long enough time horizon I think we’ll find that this was the exception, not the rule.
Second, this strategy is very subject to curve-fitting because I’m using monthly data and so few data points (n=250). But I would hedge that statement by saying this general principle worked using multiple approaches (not just EMA), multiple lookback periods (not just 6 months), etc. so I think the concept is pretty robust.
Is the TED outstanding as a technical indicator? Definitely not. But it does have its purpose and will give us something to laugh at the next time the talking heads give an increasing TED as a reason to sell.
[Edit: click for a summary of all four TED spread related posts in this series]
Happy Trading,
ms
To stay up to date with what’s happening at the MarketSci Blog, we recommend subscribing to our RSS Feed or Email Feed.
Filed under: Trading Strategies, Treasuries & Interest Rates | 8 Comments
In previous posts you noted that your strategy got smoked during recent market volatility.
Do you think there could be some benefit in looking at some logic of ‘if TED spread > n then cut position size by xx%?’
It may be worth looking into. There are so many different position sizing routines one can add to even a mediocre strategy that help out tremendously.
Also, I have been poking around a lot more with building more adaptive strategies & overnight vs. daytime gains and so far the results are promising.
Regards,
Eric
RE to Eric: Long time no talk. I think it’s possible, but I think there’s a lot more direct, straight-forward ways to accomplish the same thing. The problem I think is that LIBOR is only going to reflect very specific types of market stresses.
Here was my approach: http://marketsci.wordpress.com/2008/10/15/a-new-approach-for-coping-with-abnormal-markets-shades-of-grey/
Good to hear from you,
michael
P.S. would love to see the output of your work when you get to a stopping point.
And how do the intra-month drawdowns look?
RE to MW: this post generated quite a bit of interest so I’ll do a follow up with intra-month stats in the coming week. More to follow.
Thanks,
michael
What is the exit criteria for this system?
RE to Tim: when the entry criteria are no longer met – i.e. when the EMA is falling. michael