Dual Momentum Strategy: How Relative and Absolute Momentum Work Together, With a Worked Example

A dual momentum strategy asks two questions at every month-end instead of one. First, which of my candidate assets has done best over the past year? Second, has that winner actually beaten a risk-free return like a Treasury bill? Only if the answer to the second question is yes do we stay in equities. If it’s no, the money sits in safer debt until the trend comes back. That second filter is what separates dual momentum from plain “buy what’s been rising” investing, and it’s the part most people skip. Below we walk through both legs, how Gary Antonacci frames them, how you might adapt the idea to Indian index funds, and a worked example with illustrative numbers.

Two kinds of momentum, two different jobs

Momentum investing, in the textbook sense, means buying securities that have had high returns over the past three to twelve months and selling the ones that have done poorly over the same window. Wikipedia’s summary of the research points to Narasimhan Jegadeesh and Sheridan Titman’s 1993 study, which found that buying recent winners and selling recent losers earned average returns of about 1% a month over the following 3 to 12 months, with the edge fading after 12 months.

That is relative momentum: you compare assets against each other and pick the stronger one. It says nothing about whether the stronger one is going up at all. In a falling market the “winner” might simply be the asset losing money more slowly.

Absolute momentum fixes that blind spot. Instead of comparing an asset with its peers, you compare it with its own past and with cash. Antonacci’s FAQ on Optimal Momentum explains that his model subtracts the Treasury bill return from the stock index return over the lookback period. What’s left is the excess return. If that number is negative, his reasoning is blunt: there’s no reason to carry stock market risk when you can earn more without any downside risk.

Put the two together and you get a simple rule set. Relative momentum chooses which equity asset to own. Absolute momentum decides whether to own equities at all.

Relative momentum tells you which horse is ahead. Absolute momentum tells you whether the race is worth running this month.

How Antonacci’s original model is wired

Antonacci’s public model, which he calls Global Equities Momentum or GEM, is built around a few deliberate choices. All of the points below come from his FAQ page.

• A 12-month lookback. He says a 12-month lookback has more out-of-sample validation than any other period, tracing it to Cowles and Jones in 1937 and the Jegadeesh and Titman study of 1993.
• Monthly checks, not daily ones. GEM adjusts positions at the start of each month. His argument for not checking more often is that over short periods stocks tend to bounce back from moves rather than trend, so frequent checks invite whipsaws.
• A US versus non-US stock comparison. The relative leg compares US stocks with the MSCI ACWI ex-US index of other markets.
• Bonds as the parking spot. When the absolute filter says “out”, GEM moves to aggregate bonds. He notes that government bonds or Treasury bills can be substituted with only a modest drop in expected return.
• Order of the two filters. His book shows both orders. With only US and non-US stocks, he says it’s better to apply absolute momentum first, using the S&P 500 as the test because the US leads world equity markets. With a wider mix of assets, applying relative momentum first can be easier.

He’s also candid about the weak spots. The same FAQ lists re-entry lags after a bear market ends, tracking error against the index, and the biggest risk of all: investors losing patience and abandoning the rules at the wrong time.

Adapting the dual momentum strategy to Indian investors

Nothing in the logic is tied to the US market. What you need is two or more equity “lanes” that behave differently, plus a safe asset to fall back on. For an Indian investor working with index funds or ETFs, one sensible set-up looks like this:

• Lane 1: a large-cap index fund tracking the Nifty 50.
• Lane 2: a midcap index fund, which tends to move on a different rhythm from the large caps.
• Safe asset and hurdle rate: Treasury bills. The RBI’s FAQ on government securities explains that T-bills are short-term debt instruments issued by the Government of India in three tenors (91 day, 182 day and 364 day) and that they’re issued at a discount and redeemed at face value. A 12-month return on T-bills, or on a fund that holds them, is a natural hurdle for the absolute test.

India also has an index-provider version of the momentum idea. NSE Indices describes the Nifty200 Momentum 30 index as tracking 30 high-momentum stocks across large and mid caps, with each stock’s momentum score built from its 6-month and 12-month price returns, adjusted for volatility. That is a relative momentum stock-picking index. It doesn’t have an absolute momentum switch that takes it to cash, so it’s a cousin of dual momentum rather than the same thing. If you want the broader picture of stock-level momentum in India, our guide to momentum investing in India covers it.

One more thing to settle up front is which filter you apply first. With just two equity lanes, we’d follow Antonacci’s lead and check the broad market first (the Nifty 50 lane against T-bills), then pick between the lanes. If you add more lanes, say a gold ETF or an international fund, running relative momentum first is cleaner.

A worked example, month by month

The numbers below are hypothetical. They aren’t actual index returns. They’re there to show the mechanics, so focus on the decisions, not the figures.

Say we start with ₹5,00,000 and follow a relative-first version of the rules. On the last trading day of each month we note the trailing 12-month total return of each lane and of T-bills.

Month 1: midcaps lead, and they clear the hurdle

Relative test: midcaps beat large caps, 14.0% against 9.0%, so midcaps are the candidate.

Absolute test: the midcap excess return over T-bills is 14.0% minus 6.5%, which is 7.5 percentage points. It’s positive, so we stay in equities.

Action: the full ₹5,00,000 goes into the midcap fund for the coming month.

Month 2: the leader changes, but nothing beats cash

Relative test: large caps now lead, 4.0% against 2.5%.

Absolute test: the large-cap excess return is 4.0% minus 6.4%, which is minus 2.4 percentage points. Negative.

Action: we exit equities entirely and move to T-bills or a fund holding them. Notice what a pure relative momentum investor would have done here: switched from midcaps into large caps and stayed fully invested in a market where neither lane was beating a government bill. The absolute filter is what pulls us out.

Month 3: a close call

Now suppose large caps show 6.9% over 12 months, midcaps 6.6% and T-bills 6.4%. Large caps lead by a hair and beat T-bills by half a percentage point. The rules say get back into large caps. This is the uncomfortable part of any rules-based system. A thin margin is still a signal, and second-guessing it is how most people end up with a strategy that’s neither systematic nor discretionary. Antonacci’s FAQ mentions that when switching signals are close, he has tested splitting money equally between the two choices and found the results about the same as following the signal exactly. That’s one way to soften the jolt of a marginal call without abandoning the rules.

The rules will sometimes move you on a margin so thin it feels silly. Following them anyway is the whole point of having rules.

Where dual momentum disappoints

Every strategy has a bill that comes due, and it helps to know this one’s in advance.

It is late at both turns. A 12-month lookback means the absolute filter only flips after a decline has been running long enough to drag the trailing return below T-bills. You’ll usually take part of a fall before exiting. Coming back is late too. Antonacci himself flags re-entry lags when a new bull market starts after a bear market.

Choppy markets hurt. If the market keeps hovering around the T-bill hurdle, the rules can flip you in and out several months in a row. Each round trip costs money and nerves.

Switching isn’t free in India. Moving out of an equity fund can trigger capital gains tax and, depending on the scheme, an exit load. A strategy that looks clean on a spreadsheet can lose a chunk of its edge once you account for these. Read the scheme documents for every fund you plan to use, and think about whether the strategy fits better in a tax-advantaged account if one is available to you.

Momentum has crashed before. Wikipedia notes that momentum strategies carry the risk of occasional large crashes, citing a fall of 73.42% in three months in 2009. Antonacci argues that momentum crashes come from the short side of long/short momentum portfolios, when beaten-down stocks rebound sharply off a bottom, and that a long-only approach like his doesn’t hold those short positions. Even so, it’s a reminder that momentum isn’t a smooth ride.

Turnover eats returns. The same Wikipedia article notes that momentum strategies tend to have high turnover, and some researchers argue transaction costs can wipe out momentum profits. A monthly, index-level version like the one above trades far less than a stock-picking momentum portfolio, and Antonacci notes that the index ETFs he usually uses are more scalable than individual stocks.

Practical rules if you want to run it

1. Write the rules down before you start. Lanes, safe asset, lookback, check date and order of filters. If it isn’t written, you’ll change it the first time it hurts.
2. Use total returns where you can. Comparing a price index that excludes dividends with a T-bill return that includes all of its yield tilts the test against equities.
3. Pick one check date and stick to it. The last trading day of the month is common. Checking mid-month “just to see” defeats the purpose.
4. Keep lanes genuinely different. Two funds that track nearly the same stocks give the relative test nothing to choose between.
5. Size it as part of a plan. Dual momentum is one return stream. Antonacci’s FAQ itself argues for combining approaches that don’t move together. Our piece on risk management in trading covers position sizing and drawdown limits that apply here too.
6. Compare it with alternatives. If you’d rather rotate between sectors than between market-cap lanes, our sector rotation strategy guide uses a related idea at a finer level.

Quick summary

A dual momentum strategy stacks two filters. Relative momentum picks the stronger equity lane over the lookback, which Antonacci sets at 12 months. Absolute momentum then checks whether that lane has beaten Treasury bills, and if it hasn’t, the money moves to safe debt. Checks happen once a month. The method’s appeal is that it can step aside during long declines. Its costs are late exits and re-entries, whipsaws in sideways markets, and in India, tax and exit loads every time you switch. If you can live with those and follow the rules without tinkering, it’s one of the more disciplined ways to use momentum.

Disclaimer: This article is for educational purposes only and is not investment advice. The worked example uses hypothetical figures, not actual returns of any index or fund. Past patterns in momentum research don’t guarantee future results. Check scheme documents and current tax rules, and consult a SEBI-registered investment adviser before acting on any strategy.