The dual momentum trading strategy by Gary Antonacci, what is that about? As the Oracle of Omaha, Warren Buffett, once said: “Trying to time the market is the number one mistake to avoid.” It is almost impossible to consistently time the market — you will either be buying or selling too late or too early rather than at the right time, which is why many professional investors advise against that. One method that has offered a way to spot and get into the right trend is Dual Momentum.
The dual momentum trading strategy by Gary Antonacci is a method of investing that selects only assets that have outperformed their peers over a given time and also making positive returns. It is based on the idea that an asset with a superior relative momentum and a positive absolute momentum would continue to perform until another outperforms it. Thus, it is a sort of trend strategy.
- A comprehensive library of trading systems
- 3 Momentum Trading Strategies (Backtests, Setups, Rules, And Indicators)
To help you understand the topic, we will discuss it under the following headings:
What is dual momentum?
Dual momentum investing strategy uses two kinds of momentum to determine which security to buy and when to do that.
The strategy got its name from the fact that it uses two types of momentum in its analysis. It compares the current momentum of two or more financial securities and chooses the one with the greatest momentum and then compares that with what it was in the past.
That is to say, the dual momentum approach seeks to invest in an asset only if it is performing better than its peers over a given period and has a positive (upward) momentum at the same time. It, therefore, does not aim to buy the best among losers; it only aims to buy the best among the performers.
On the other hand, in markets such as futures where you can also go short as easily as you can go long, the dual momentum strategy can also be used to identify and short the worst-performing asset that has a negative (downward) momentum at that momentum — the weakest of the weak.
Plenty of academic research has proven the momentum effect in stocks for short-term periods ranging from one to 12 months. For longer and shorter periods it has not proven to work well. For example, stocks that have performed well over the last 6 months tend to continue to perform well the next 6 months.
To better understand the concept of dual momentum, we need to understand what momentum means, the types of momentum, and the dual momentum formula.
The momentum formula: what does momentum mean?
Momentum, as it’s used in the momentum oscillator in many trading platforms, is also known as the rate of change (ROC), which measures the amount that a security’s price has changed over a given period. It is often expressed as the percentage change in the current price compared to a specific period in the past.
The formula is given as follows:
Momentum = [Current Price / Price at a given period in the past (usually 14)] x 100
As you can see, the momentum is achieved by dividing the current price by the price of at a specific previous period. This gives you the proportional rate of change. To get the percentage rate of change, you multiply the quotient by 100. For the momentum (ROC) indicator seen on different trading platforms, this gives an indicator that oscillates around 100, such that values less than 100 indicate negative or downward momentum (decreasing prices) while values more than 100 indicate positive or upward momentum (increasing prices).
Types of momentum
The formula we explained above is for one type of momentum known as the absolute momentum, but there is also another concept of momentum known as the relative momentum. Basically, there are two types of momentum:
- Absolute momentum
- Relative momentum
This is the momentum of an asset relative to itself. It is simply the percentage rate of change of the price of the asset over a specified period. If the asset has a positive change over the period of interest, the absolute momentum would be positive, and the greater the change, the greater the momentum and vice versa.
This type of momentum compares the returns of an asset over a given period to those of other assets. In other words, it compares the absolute momentum of one asset to that of another asset or even those of many assets such that the asset with a higher momentum than the others is said to have positive relative momentum.
The dual momentum formula
It is a bit difficult to express the dual momentum concept in a simplified formula that a newbie can easily understand. However, we can simplify the concept as follows:
Dual Momentum = Absolute Momentum of Asset 1 – Absolute Momentum of Asset 2, if both absolute momentums are positive or negative.
And, the interpretation would be as follows:
- Go long on asset 1 if dual momentum is positive and both assets have positive absolute momentum (NB: Asset 1 represents the asset with a higher absolute momentum).
- Go short on asset 2 if the dual momentum is negative and both assets have negative dual momentum (NB: Asset 1 represents the asset with a higher absolute momentum).
The formula is just to get a positive relative momentum in a broadly positive market or a negative relative momentum in a broadly negative market. With that, you will be buying on the best of the positively performing assets and selling the worst of the negatively performing assets.
How dual momentum works
The dual momentum strategy works by targeting the assets with the most momentum in a particular market situation. The idea is to select assets that have historically outperformed other assets and are also themselves in positive momentum — these would be the assets to buy, especially in a bull market. On the flip side, if you want to go short in a downward market, you should aim to select those assets that have both negative absolute and relative momentum.
Here’s the explanation: When the general market is bullish, you want to be buying the best-performing securities, and when the market is bearish, you want to be selling the worst-performing securities. So, you aim to buy securities with both positive absolute momentum and positive relative momentum, while you sell securities that have both negative absolute momentum and negative relative momentum.
Mind you that it is possible to assets that have positive absolute momentum but negative relative momentum — for example, when the market is generally going up, the market laggards can have positive momentum but perform poorly when compared to their peers.
You don’t want to buy these laggards; instead, you want to buy the leaders with positive relative momentum. On the other hand, there can be assets with positive relative momentum and negative absolute — for instance, when the market is generally going down (a bear market), some assets with negative absolute momentum would have positive relative momentum because they are losing less than others. But you surely won’t want to be buying them because they are also losers; instead, you will look to short the worst losers if it’s a market you can easily go short.
The first step in applying the dual momentum approach is to find assets with positive absolute momentum. So, you have to calculate the absolute momentum of individual assets of interest over a specific period you have chosen. You get this by dividing the current price of each asset by what it was at a specified period in the past. If an asset is trading at a higher price than it was in the past, its absolute momentum would be positive, and you should consider it for a long position after checking the relative momentum. However, if it’s trading lower, its absolute momentum would be negative, making it a candidate for short selling.
The second step is to compare the assets with positive absolute momentum against one another to get the relative momentum. You also compare assets with negative absolute momentum against one another to get their relative momentum. Among assets with positive absolute momentum, the ones with the highest relative momentum should be selected for buying. For those with negative absolute momentum, the ones with the lowest relative momentum (biggest negative value) should be for short positions. See the table below:
In essence, the dual momentum approach forces you to buy assets that are both going up and outperforming their peers, while selling those ones that are both going down and performing worse than the other losers. For the equity market, the dual momentum strategy performs better for the long side because the equity market tends to have a long-term upward bias.
If you want to take advantage of the short side too, you should look for a market that offers easy access to both the long and short sides, such as futures and forex markets. However, you may not want to use the strategy in forex because long-term trends are hard to find in the currency market these days.
The dual momentum strategy: other things you need to know
There are other things you should know about the dual momentum strategy. These are some of them:
What is the origin of the dual momentum strategy?
The origin of the dual momentum strategy can be traced to the seminal work of Jagadeesh and Titman (1993), which showed that using relative momentum to make investment decisions could provide profitable trading opportunities that are robust enough to be exploited using certain parameters.
Their work indicated that the returns of relative momentum outperformed benchmark returns, but the volatility inherent in the strategy is only marginally better than that of the benchmark. So, some active investors and hedge fund managers don’t believe that the reward is big enough to justify the risk.
However, in 2012, Gary Antonacci published a book titled, “Dual Momentum Investing: An Innovative Strategy for Higher Returns with Lower Risk”, where he explained a simple but highly effective extension of the relative momentum approach. He added another aspect into the concept by bringing the asset’s absolute momentum into the analysis. What he found was that by combining the two types of momentum, it was possible to enjoy the rewards of the relative momentum approach while greatly reducing the volatility that’s inherent in the approach.
How to identify the dual momentum signals
There is no special way to identify the dual momentum signal except to follow the main idea in the concept to buy the strongest among the strong and sell the weakest among the weak. The flow chart below simplifies the entire process.
You start by finding the absolute momentum of the assets of interest, which could be securities in related sectors of the same market or different markets like equities and bonds. Then, compare their momentum to choose the best performers among those securities that have positive absolute momentum. These are the ones showing a long signal.
For markets where you can safely go short, compare all the securities showing negative momentum and select the worst performers, as they are the ones showing a short signal. There is no hard and fast rule about when to enter a trade, but there tend to be more long signals when the general market is bullish, and more short signals in a bearish market environment. However, in a strong
Examples of the dual momentum strategy
Let’s discuss two scenarios to show how the dual momentum strategy can be applied in real-life situations:
First example: S&P 500 Index vs. Gold
In our first example, let’s say you are monitoring the S&P 500 Index and the commodity market — say gold, for instance. Assuming the S&P 500 Index was trading at 3025.52 three months ago and has made a 10% increase in price over that period. Gold, on the other hand, has made a gain of about 5% over that same three months.
Thus, in terms of absolute momentum, the S&P 500 Index would be reading an absolute momentum of 110%, while the absolute momentum of gold would be 105%. Comparing the momentum in the two assets, it’s clear that the S&P has greater momentum than gold at the moment, so the relative momentum is positive for the S&P.
Since the S&P Index has both positive relative momentum and positive absolute momentum, it is the right asset to buy at the moment since it is the better of the two assets. The belief is that the asset with a higher momentum is more likely to continue gaining more momentum, with prices rising more in the nearest future. However, you don’t buy the S&P 500 Index directly; instead, you buy an ETF that tracks it, such as the SPY ETF (SPDR S&P 500 Trust ETF).
We made a backtest of the SPY vs. GLD ratio (momentum):
Second example: Crude Oil and Natural Gas Futures
For our second example, we will consider two commodities in the futures market where you can easily trade on the short side no matter your account size. Now, let’s say you are monitoring crude oil and natural gas futures, and over the last one month, both assets have been on the decline. Assuming crude oil made a 10% decline in price over that period, and natural gas, on the other hand, made a decline of about 5% over that same period.
Using those values to calculate the absolute momentum of those assets, crude oil would be reading an absolute momentum of 90%, while the absolute momentum of natural gas would be 95%. Comparing the momentum in the two assets, it’s clear that crude oil has a greater downward momentum (negative momentum) than gold at the moment. In other words, the relative momentum between those two assets would be negative for crude oil.
With crude oil showing both negative relative momentum and negative absolute momentum, it is the right asset to short at the moment because it is the worse of the two assets. This is in line with the belief that the asset with a greater downward momentum is more likely to continue to decline in the nearest future.
How has the dual momentum strategy performed in the past?
Interestingly, the dual momentum strategy, despite using a relatively simple approach, has performed remarkably well over the years, as Gary Antonacci (who created the strategy) demonstrated with his Global Equities Momentum (GEM) model, which holds U.S. or non-U.S. stock indices when stocks are strong and uses bonds as a safe harbor when stocks are weak.
Gary presented his GEM results from as far as 1950 up to 2018 and compared them to a global asset allocation (GAA) benchmark of 45% in U.S. stocks, 28% in non-U.S. stocks, and 27% in 5-year bonds.
Gary stated that these percentages represent the amount of time GEM spent in each of these markets and may also be considered to represent a typical global asset allocation portfolio. However, these backtesting results are simulated and hypothetical. They should never be considered indicative of future results, as they do not represent returns that any investor actually attained in the past. You should know that indexes cannot be directly traded, so they do not reflect management or trading fees when used in backtesting.
The results show that the correlation in monthly returns between GEM and GAA is 0.60, but between GEM and the S&P 500, the correlation is 0.50. Since, on average, there were only 1.5 trades per year with GEM, transaction costs must have been minimal.
Interestingly, these results are still ongoing and are updated monthly on the Performance pages of Gary’s website.
The chart below is taken from Allocatesmartly.com and shows how the strategy has performed relative to the traditional 60/40 portfolio:
Allocatesmartly.com goes on to show the performance metrics and statistics:
The drawdowns are low and relatively short-lived. We know that most traders abandon a strategy when drawdowns approach 20% or more, thus this strategy might be interesting as a substitute for the traditional buy & hold of stocks.
Dual momentum strategy – backtest and performance
Let’s make a backtest of Antonacci’s dual momentum strategy. The trading rules are like this:
- Did S&P 500 outperform US Treasury Bills over the last 12 months?
- If the answer is no, then buy US Treasury Bills.
- If the answer is yes, did S&P 500 outperform the world index (ex. USA) over the last 12 months?
- If S&P 500 outperformed the world index, then buy S&P 500.
- If S&P 500 underperformed the world index, then buy the world index (global stocks).
These trading rules are asked at the close of every month and repeated 12 times per year. Thus, this is a momentum strategy with monthly rebalancing, and we hold one asset at a time. Rinse and repeat monthly.
For the backtest above, we used the following ETFs: SPY for S&P 500, EFA for global stocks (world index ex. USA), and AGG for Treasury Bills (not an optimal proxy, but serves our purpose). Dividends are included and reinvested.
The equity curve of the strategy looks like this:
The annual return was 6.75%, and the max drawdown was 30%. This result is better than the world index, but it failed to beat S&P 500, which compounded at 9.2%, however, with a max drawdown of about 55%.
Investing with the dual momentum strategy
Surely there are many ways you can implement the dual momentum concept when trading/investing. You can use the approach to trade individual assets, such as stocks and securities on the futures markets. However, Gary’s research shows that momentum works best when used to invest in ETFs, especially the ones that track geographically diversified equity indexes.
Implementing the dual momentum concept with individual assets
You can tweak this concept and use it to select the stocks to invest in different market conditions. For example, during a bullish market, you can use the approach to screen stocks such that you only select the ones with the greatest momentum and invest in those ones.
Similarly, in a bear market, you can use the concept to select the securities with the biggest downward momentum and go short on them.
Implementing the dual momentum strategy with ETFs
The best way to implement the dual momentum strategy is with ETFs. Gary described a modular approach where, every month, you compare two related sectors or two parts of a single sector and select the one that performed better over a certain period — usually the preceding twelve months. You buy the better performer if it has positive absolute momentum, but if it has negative absolute momentum, you opt for treasury bonds or investment-grade bonds.
These are examples of the markets that Gary suggested for this approach:
- Bonds: credit bonds & high yield bonds
- Equities: US equities & international equities
- Economic stress: gold & treasury bonds
- REITS: mortgage reits & credit reits
If well implemented, the approach has the potential to yield some really good results. However, Gary indicated that the dual momentum strategy tends to underperform during strong bull markets or when the market rebounds strongly, but with its ability to weather bear markets, the strategy has outperformed in the long term.
Using the dual momentum concept in sector rotation
The dual momentum concept can also be applied to the sector rotation approach. In this case, you may create a universe of ETFs that represent various sectors, regions, or asset classes and then rank the sectors/regions/asset classes according to their absolute momentum over a chosen period.
You can then decide to buy the best three ETFs with positive absolute momentum. After a certain period, you rotate again into the best three at that moment. So, you are basically leveraging the dual momentum approach in applying the sector rotation strategy.
Other momentum strategies
We have backtested hundreds of trading strategies since we started the blog in 2012. Some of them have been momentum strategies:
- The momentum day trading strategy in the S&P 500 (SPY) during the last hour
- Is Meb Faber’s momentum/trend-following strategy in gold, stocks, and bonds still working?
- S&P 500 Momentum Strategy – As Simple As It Gets (Backtested)
- Monthly momentum in SPY and EEM
- Monthly momentum in SPY and TLT (rotation strategy)
- A monthly momentum strategy in ETFs (sector rotation in EEM, SPY, TLT)
- Trend following and momentum strategies on bitcoin (crypto) – capturing the trend (free bitcoin trading strategies)
The benefits of using the dual momentum strategy
There are two key benefits of the dual momentum strategy. The first one is that it offers a better result than the buy and hold strategy. The backtested results displayed above shows that the dual momentum approach performed better.
The second key benefit is that it comes with less volatility. Its worst drawdowns were not as bad as that of the benchmark.