Why No Bond Sell Signal?

(This is an excerpt from Wednesday's blog for Decision Point subscribers.)

As of 11/10/2010 Bonds are on a Trend Model Neutral signal. A subscriber asks: "Bonds have had a massive pullback during the last few weeks, but your bond posture has only changed from bullish to neutral. Out of curiosity, if you are willing to share, what has to happen to move it to a bearish posture?"

ANSWER: We receive this question regarding our Trend Model often. First, signals on the Trend Model are mechanical and can be buy, neutral, or sell. A buy signal can be generated anytime the 20-EMA crosses up through the 50-EMA. A sell signal is generated only when the negative 20/50-EMA crossover occurs when the 50-EMA is below the 200-EMA, which means that the price index is in a long-term bear market. If the negative crossover occurs when the 50-EMA is above the 200-EMA, a neutral signal is generated, which was the case on November 10. This is a conservative approach to downturns that is built into the model. It allows us to sidestep declines without going short prematurely (i.e. before a long-term bear market has been objectively declared).

Below is the Bond chart. You can see that we had a buy signal back in April, and then the negative crossover of the 20/50-EMAs occurred last month when the 50-EMA was above the 200-EMA, generating a neutral signal. The 50-EMA is still above the 200-EMA, so by definition, bonds are still in a long-term bull market. We don't know the ultimate outcome, but it is quite nice to at least be out of the bond market as it unravels.

On the Dollar chart below you can see the neutral signal, followed by the 50-EMA crossing below the 200-EMA generating a LT sell signal. We then had the recent buy signal generate. Should the dollar break down from this point, we'll have a negative crossover with the 50-EMA below the 200-EMA and a sell signal will be generated.

A neutral signal could be treated as a sell signal on a discretionary basis by more aggressive traders or those that can quickly close out of a short position that is going against them. The reason we take a more conservative approach in the model design is that being short at the wrong time can seriously affect performance of our models versus performance of the price index. If we are on a buy signal while the market declines, our performance exactly matches the price index. However, if we are on a sell signal while the market rallies, say +5%, we not only lose that gain, we also lose -5% on our short position, which puts us 10% behind index performance.

For a more detailed discussion, check out our learning center article on the Decision Point Trend Model.

* * * * * * * * * * * * * * * * * * * * *

Technical analysis is a windsock, not a crystal ball.

* * * * * * * * * * * * * * * * * * * * *

Decision Point's Timer Digest Rankings

Copyright 2010 Decision Point. Nothing herein should be construed as an offer or solicitation to buy or sell any security. Past performance does not indicate future results.

About the Author

President
cswenlin [at] decisionpoint [dot] com ()
randomness