This article deals with diversification and momentum. It brings the two concepts together in a non-traditional manner. Our intent is to see if value can be created where:

Value is defined as higher return for the same amount of risk or lower risk for the same amount of return.

A brief review of where we have been follows.

Diversification

We looked at traditional diversification represented by a variety of weightings of two (diversified) assets — stocks and bonds. The results showed risk reduction but with pretty heavy penalties to returns.

A subsequent article looked at expanding the asset classes beyond generic stocks and bonds to see what could be accomplished. This more diversified portfolio improved over the previous work but still seemed unexciting.

After that, a simple moving average (SMA) was introduced. This technique, not considered diversification by most, produced impressive results. It is included here because it represented two assets, one a diversified collection of US Stocks and the other a diversified collection of US Bonds. In that sense, it was like the first attempt at traditional diversification mentioned above. What differed here is that you were only in one at a time, allowing market performance to dictate which and for how long.

The simple moving average (SMA) signaling switch produced superior results, both in terms of return and risk to the traditional approaches to diversification. One might think of it as a form of serial diversification.

Momentum

The last article looked at momentum as a way to invest. That article pointed out that a moving average works because trends, once begun, tend to last longer than Efficient Market Theory (EMT) predicts. In that sense, the SMA switch could be considered a quasi-momentum tool as well as a serial diversification tool.

The purpose of the previous article was to explore using past returns, the traditional momentum measure, to try to produce superior results. Higher returns tend to persist for longer than EMT predicts. Ten assets were ranked based on momentum and then the highest ones were purchased. Assets were held for a month and then the process repeated.

Backtesting results were dramatic both in terms of risk reduction and return enhancement. Results superior to all previously tested strategies were achieved. The optimum number of assets to select from the ten availablewas two, although all tested provided excellent outcomes.

Where Are We?

At this point, several basic approaches to investing have been explored. The best results were achieved (in the previous article) using momentum.

The best results were obtained from a two asset portfolio. It is doubtful many investors would be willing to trade a two stock portfolio, even if the two were themselves reasonably diversified.

Most investors are wisely grounded by the old adage of not “putting all your eggs into one basket.” That advice will be challenged, but not necessarily contradicted by future articles in this series.

From here we shall look at the effects of combining different methods of investing.

This first article will explore the combination of two previously investigated methods — Momentum and an SMA switch.

Diversification and Momentum

High volatility and high drawdowns, can be dampened by diversification. Often this diversification is costly in terms of what it does to returns.

The best returns seen thus fare are from the momentum selections. These also produced lower risk than traditional diversifications methods. The SMA switch also achieved more favorable outcomes than traditional diversification or buy and hold.

In this article, the effects of combining these two methods will be reviewed. Can we create value by creating a portfolio from two distinct methods of selecting holdings? Can we improve risk or return or both?.

The Test

The following represents holding part of your funds in the SMA model and the rest in the simple momentum selection model. Both models were discussed in detail in previous articles.

A couple of issues needed to be addressed to accomplish the combination:

1. Should the SMA be applied to only the one model or to both? The SMA switch was only applied to the market index. The rationale was that it was not used in the original momentum tests and would alter the outcomes from a model we wished to include based on its own performance.
2. What should be the proportions between the two different approaches? Three different proportions were used in the backtesting process as can be seen below. There is no “right” mix. The choice is a personal and empiric one.

Tests were performed on data from 1998 through January 2019. The SMA Switching Index (SMA) was used as was the results of momentum selection (MOM) of 2 assets from ten used in the last article

Three different weightings were used in the testing:

• Portfolio 1: 80% SMA; 20% MOM
• Portfolio 2: 60% SMA; 40% MOM
• Portfolio 3: 40% SMA; 60% MOM

The Results

The results of each test and the performance of the Vanguard 500 Index (a buy and hold strategy) are shown below:

DIVERSIFICATION AND MOMENTUM TESTS

The Interpretation of the Tests

The results clearly suggest there is value in combining the two approaches. Each of the three combinations outperforms the buy and hold strategy on both risk and return.

One might have expected the SMA portfolio to be less volatile than the 2 asset MOM portfolio. However, as the proportion of MOM increases in the combinations above, drawdown is virtually unaffected and return improves. This result seems counterintuitive, at least at first.

All portfolios beat the buy and hold strategy of the Vanguard 500 index and all do so handily and in every respect, especially with respect to maximum drawdown.

Caveats

These results were rather dramatic, but they still presumably do not satisfy risk-averse investors. An investor’s portfolio, following this strategy, now consists of three assets at any one time, a stock or bond index and two momentum selections.

Three assets as a portfolio is apt to be unsettling, however, perhaps it should not be. One of the assets (the stock or bond index) is always a fully diversified asset. The other two assets (from the momentum selections) are also diversified selections, generally narrower than total markets. Gold which is a “run for the hills” asset is an exception. Four of the assets in the Mom listing of ten are cash or near-cash selections, allowing the momentum selection easy access to safety selections and the equity assets are diversified, at least within defined classes.

For most, three assets will be off-putting, although perhaps it should not be given the diversified nature of the assets. Diversification is good up to a point. Beyond that point, diversification for the sake of diversification is a waste of time and money. (Statistical analysis of stocks from different industries suggests that diversification beyond ten stocks provides little to no additional risk reduction.)

Additional methods to achieve value results (in terms of the risk-return trade-offs) will be explored and reported on in subsequent articles.

Good luck and successful investing.