Blog - Investing Notes

October 23, 2008 - Update: The Decision Moose Asset Allocation Framework

Reader Bob Tait of Houston suggested a review of the free Decision Moose asset allocation framework of William Dirlam. "Decision Moose is an automated framework for making intermediate-term investment decisions." Decision Moose focuses on asset class momentum, as augmented by monetary policy, exchange rate and interest rate indicators. Its signals tell followers when to switch from one index fund to another among nine encompassing a broad range of asset classes, including equity indexes for several regions of the globe. The trading system is a long-only approach that allocates 100% of funds to the index "having the highest probability of price appreciation." The site includes a history of switch recommendations since the end of August 1996. Mr. Dirlam provides background on the site and the Decision Moose framework via FAQs. To evaluate the framework, we assume that the 51 switches and trading returns are as described (out of sample, not backtested) and compare the returns to those for the S&P 500 index over the same trading intervals (based on all positions closed as of 10/17/08). We find that:

The following chart summarizes results by trade for all 51 closed Decision Moose trades and comparable S&P 500 index trades. Results exclude dividends and trading frictions.

• Decision Moose generates an average of about four trading decisions per year.
• For the Decision Moose (S&P 500 index) trades, 92% (59%) generate positive returns. Decision Moose returns beat the comparable S&P 500 index returns 69% of the time.
• The largest loss on any one trade for Decision Moose (S&P 500 index) is -5.1% (-22.5%).
• The average return per trade for Decision Moose (S&P 500 index) is 7.3% (1.0%), with standard deviation 15.0% (7.7%).
• Over just the last three years, the average return per trade for Decision Moose (S&P 500 index) is 3.0% (-1.1%), with standard deviation 3.6% (6.7%).

Does Decision Moose investing have a hedging effect?

The following scatter plot presents a different perspective on the same data. The data point at the upper left of the plot is the Trade 18, an extraordinary gain in gold shares for Decision Moose. Visual inspection indicates that Decision Moose generates no large losses and often generates gains when the S&P 500 index has losses.

The Pearson correlation between the two series is 0.06 and the R-squared statistic is 0.00, indicating no relationship and implying good hedging. However, excluding the Trade 18 outlier increases this correlation to 0.42 and the R-squared statistic to 0.18, indicating some positive link between Decision Moose results and the performance of U.S. equities. The large change in correlation caused by excluding just one data point says the sample is somewhat small for deciding the strength of the relationship.

In his FAQs, Mr. Dirlam suggests that Decision Moose trading is best suited to a reasonably large tax-deferred account to minimize the impacts of trading friction and taxes. He also offers guidance on the type of investor for whom Decision Moose is suitable.

In summary, the Decision Moose asset allocation framework may offer investors a way to beat buying and holding broad U.S. equity indexes by occasionally trading to the "hottest hand" from a broad set of asset classes.

It is plausible that investing across asset classes and around the globe amplifies and extends momentum by exploiting the reluctance of some (most?) investors to go outside their investing comfort zones. However, this amplification of momentum might diminish with time as investors think more globally.

For related research, see Blog Synthesis: Momentum Investing/Trading and Blog Synthesis: Reviews of Books and Web Sites. See especially our blog entries of 2/27/07 and 2/15/07 on somewhat comparable momentum-based asset class trading systems.