Wednesday, July 29, 2009

The Faber Model And Inflation Pressures

Several weeks ago I presented research that improved the efficiency of the Faber market timing model for the S&P500 by some 50%. By efficiency I meant that the new and improved model made more money with less time in the market and with less draw down. The research can be found in this article, "Inflationary Pressures Are A Legitimate Concern".

The gist of the research was that stocks tended to under perform during times when the trends in gold, commodities, and yields on the 10 year Treasury bond were strong. The Faber model is a simple moving average model, yet we can improve the model's efficiency (for the S&P500) by moving to cash when (real or perceived) inflation pressures are strong as measured by a composite indicator that assesses the trends in gold, commodities, and yields on the 10 year Treasury bond.

So at the end of last month, the Faber model gave a buy signal for the S&P500, which is a monthly close over the simple 10 month moving average. Based upon the research in this article and other factors, I elected not to take this buy signal. As to my "call" on the markets, things were looking good as prices were some 5% below the original buy signal from the Faber model, but during the week of July 17, the S&P500 reversed higher, and by the end of the month, we are now some 12% off those lows. Not surprisingly, during the week of July 10, the trends in gold, commodities, and yields on the 10 year Treasury had come off appreciably opening the way for higher equity prices.

So here comes the end of the month, the buy signal from the Faber model is in the black by about 5.5%, which means the S&P500 is up 5.5% for the month of July. Inflation pressures have come off considerably, and this can be seen in our indicator that measures the trends in commodities, yields on the 10 year Treasury bond, and gold. See figure 1 a monthly graph of the S&P500. So according to our revised (more efficient) Faber model, this is now a new buy signal as inflation pressures are low and prices are above the simple 10 month moving average.

Figure 1. S&P500/ monthly

So how do I interpret this signal? The research and premise of the research is sound - stocks do not perform well during high (real or perceived) inflationary pressures. The corollary to this is that with low inflation pressures the environment for a bullish (i.e., sustainable) run is more sound.

So how do I balance a mostly bearish bent with this bullish back drop from the revised Faber model? In other words, how do I avoid a situation like 2002 where there was a vicious head fake as the fundamentals became divorced from the technicals? The best answer I can give you is to wait for a more risk adjusted entry point.

Let me show you the MFE graph from the original Faber model, which utilizes the simple 10 month moving average solely. See figure 2. MFE stands for maximum favorable excursion. You put on a trade, and hopefully, it runs up, and then the trade is closed out for a loss or a win. So looking at the caret in the blue box in figure 2, we see that it ran up 21% above its entry point (x-axis) before being closed out for a 15% winner (y- axis). We know this trade was a winner because it is a green caret. What we see from this graph is that the current 5.5% profit from this trade does not guarantee that this trade is going to be a winner. In fact, there are at least 4 trades (inside the oval) that ran up at least 10% before reversing and giving back their gains. In other words, the strength over the past month does not indicate we are out of the woods.

Figure 2. MFE Graph/ Faber Model

Now let's look at the MAE graph from the new, upcoming signal - this is the Faber model plus low inflation buy signal. MAE stands for maximum adverse excursion,and it assesses each trade from the strategy and determines how much a trade had to lose before being closed out for a winner or loser. For example, look at the caret with the blue box around it. This one trade lost 6.4% percent (x-axis) before being closed out for a 20% winner (y-axis). We know this was a winning trade because it is a green caret. What we see from the MAE graph is that about 50% of the trades had an MAE greater than 3%, and this is to the right of the blue vertical line. It should also be noted that all the losing trades from this strategy had an MAE greater than 3%. An MAE greater than 6.5% (or to the right of the red line) always (2 occurrences) resulted in a losing trade.

Figure 3. MAE Graph/ Faber Model Plus Low Inflation

So once again: how do I interpret the signal when there is low inflation as measured by our composite indicator and prices close above the simple 10 month moving average?

1) As the trading characteristics of the original Faber model show, a 5.5% run up above the signal entry in the S&P500 does not always translate into higher prices over time.

2) In the modified model, which is triggering a buy signal at the end of this month, I have a 50% chance of getting a pullback of at least 3%. Unfortunately, there is no "free lunch" here. As the MAE graph shows, all the losing trades from the modified strategy had an MAE greater than 3%. So while waiting for a pullback to lower my risk may seem prudent, there is an increasing risk that the trade will be a bust anyway.

3) A stop loss should be set 6.5% below this Friday's closing (end of the month) price on the S&P500. In the modified model, only 2 trades out of 40 had MAE's greater than 6.5% and both were losers. A stop loss set at this level hopefully will be out of the "noise" of the market.

1 comment:

DavidDT said...

Great work Sir
I think we hit that "sell time" you referred to 2 weeks ago
http://tinyurl.com/l99qb9
Thank you for your hard work