Momentum

Momentum is the tendency of investments to persist in their performance. Assets that perform well over a 3 to 12 month period tend to continue to perform well into the future. The momentum effect of Jegadeesh and Titman (1993) is one of the strongest and most pervasive financial phenomena. Momentum investment strategies have been mostly applied to equities (see momentum in equities), however there is large evidence documenting momentum across different asset classes. Typical strategy consists of a universe of major indices on equity, bonds, real estate and commodities. The aim is to keep long only portfolio where an index with positive past 12 month returns is bought and negative returns sold. A well-known example of trend following momentum strategy is from Faber (2007). He creates 10 month moving average for which assets are sold and bought every month based on price being above or below the moving average. Using a 100 years of data, Faber claims to outperform the market with the mean return of 10.18% , 11.97 % volatility and max draw-down of 50.29%, compared to S&P 500 return of 9.32%, volatility of 17.87% and max draw-down of 83.46%.

In general, we distinguish between absolute and relative momentum. Absolute momentum is captured by trend following strategies that adjusts weights of assets based on past returns such as relative level of current prices compared to moving averages. Relative or cross sectional momentum, on the other hand, use long and short positions applied to both the long and short side of a market simultaneously. It makes little difference whether the studied markets go up or down, since short momentum positions hedge long ones, and vice versa. When looking only at long side momentum, however, it is desirable to be long only when both absolute and relative momentum are positive, since long-only momentum results are highly regime dependent. In order to increase performance, the simple momentum strategy is expanded to capture both relative and absolute momentum creating a long short portfolio.

Various extensions to the simple strategies shown above have been suggested. For example we can deploy mean-variance optimisation to re-weight our assets to minimise the risk given return. Moreover, we can diversify the strategy by restricting the weights to different asset classes and risk factors as well as adding various risk management practices to decrease leverage during heightened volatility periods. Furthermore, taking into account the cyclicality and idiosyncratic momentum of various sub-indices to Faber’s original asset classes produces even stronger improvements to risk-adjusted returns. Unfortunately, cross-sectional strategies use high number of stocks resulting in high trading costs. Luckily, it has been found that using sectors and indices instead of individual stocks still earns similar momentum returns while having lower trading costs.

Numerous empirical studies report on benefits of extending momentum strategy across asset classes (see Rouwenhorst 1998, Blake 1999, Griffin, Ji, and Martin 2003, Gorton, Hayashi, and Rouwenhorst 2008, Asness, Moskowitz, and Pedersen 2009). For example, including commodities in a momentum strategy can achieve better diversification and protection from inflation while having equity like returns (Erb and Harvey, 2006). Foreign exchange is another asset class with published momentum effects. Okunev and White (2003) find the well-documented profitability of momentum strategies with equities to hold for currencies throughout the 1980s and the 1990s. Contrary to already mentioned asset classes, bond returns have generally not displayed momentum. However, some later evidence suggests that assorting bonds with volatility adjusted returns leads to observation of momentum. Using 68,914 individual investment-grade and high-yield bonds, Jostova et al. (2013) find strong evidence of momentum profitability in US corporate bonds over the period from 1973 to 2008. Past six-month winners outperform past six-month losers by 61 basis points per month over a six-month holding period. Last but not least, momentum has been documented in real estate with a cross-sectional momentum buy/sell strategy significantly reducing volatility and drawdown of a long only REIT fund.

An often cited benefit of momentum strategies is their sustainable performance attributed to a true anomaly rather than skewedness in the return probability distribution that is cited to be responsible for value and carry strategy. Reasons explaining the momentum anomaly include analyst coverage, analyst forecast dispersion, illiquidity, price level, age, size, credit rating, return chasing and confirmation bias, market-to-book, turnover and others.

Can We Profit from Disagreements Between Machine Learning and Trend-Following Models?

26.June 2025

When using machine learning to forecast global equity returns, it’s tempting to focus on the raw prediction—whether some stock market is expected to go up or down. But our research shows that the real value lies elsewhere. What matters most isn’t the level or direction of the machine learning model’s forecast but how much it differs from a simple, price-based benchmark—such as a naive moving average signal. When that gap is wide, it often reveals hidden mispricings. In other words, it’s not about whether the ML model predicts positive or negative returns but whether its view disagrees sharply with what a basic trend-following model would suggest. Those moments of disagreement offer the most compelling opportunities for tactical country allocation.

Continue reading

Can We Finally Use ChatGPT as a Quantitative Analyst?

30.May 2025

In two of our previous articles, we explored the idea of using artificial intelligence to backtest trading strategies. Since then, AI has continued to develop, with tools like ChatGPT evolving from simple Q&A assistants into more complex tools that may aid in developing and testing investment strategies—at least, according to some of the more optimistic voices in the field. Over a year has passed since our first experiments, and with all the current hype around the usefulness of large language models (LLMs), we believe it’s the right time to critically revisit this topic. Therefore, our goal is to evaluate how well today’s AI models can perform as quasi-junior quantitative analysts—highlighting not only the promising use cases but also the limitations that still remain.

Continue reading

Revisiting Pragmatic Asset Allocation: Simple Rules for Complex Times

30.April 2025

Pragmatic Asset Allocation (PAA) represents a portfolio construction approach that seeks to balance the benefits of systematic trend-following with the realities faced by semi-active investors (mainly taxes and lack of time to manage positions). Approximately a month ago, we ran a test and filtered asset allocation strategies from our Screener and looked for those that performed well on a YTD basis. One of those models that fared surprisingly well was the PAA model, and given the challenging market conditions so far in 2025, with mixed signals across asset classes and increased macroeconomic uncertainty, we believe it is an ideal time to revisit the PAA framework. This analysis may help clarify whether a pragmatic, rules-based approach can still hold its ground—or even outperform—in a year when many models have struggled.

Continue reading

Navigating Market Turmoil with Quantpedia Tools: A Rational Guide for Portfolio Management

7.April 2025

The recent imposition of sweeping global tariffs by President Donald Trump has triggered a sharp and sudden selloff across global equity markets. In times like these, it’s natural for panic to set in. However, as quantitative investors, our strength lies in data-driven decision-making, risk management, and maintaining discipline when others lose theirs.

Rather than reacting emotionally, the prudent course of action is to reassess the robustness of our portfolios. Are we diversified across uncorrelated strategies? Do we have components in place that act as hedges during market crises? Fortunately, the tools provided by Quantpedia can help investors, traders, and portfolio managers identify, test, and deploy crisis-resilient strategies in a structured and evidence-based manner.

Continue reading

Can Margin Debt Help Predict SPY’s Growth & Bear Markets?

5.March 2025

Navigating the financial markets requires a keen understanding of risk sentiment, and one often-overlooked dataset that provides valuable insights is FINRA’s margin debt statistics. Reported monthly, these figures track the total debit balances in customers’ securities margin accounts—a key proxy for speculative activity in the market. Since margin accounts are heavily used for leveraged trades, shifts in margin debt levels can signal changes in overall risk appetite. Our research explores how this dataset can be leveraged as a market timing tool for US stock indexes, enhancing traditional trend-following strategies that rely solely on price action. Given the current uncertainty surrounding Trump’s presidency, margin debt data could serve as a warning system, helping investors distinguish between market corrections and deeper bear markets.

Continue reading

Does the Image-Based Industry Classification Outperform?

18.February 2025

For decades, investors and analysts have relied on traditional industry classifications like GICS, NAICS, or SIC to group companies into sectors and peer groups. However, these rigid categorizations often fail to capture the evolving nature of businesses, especially in an era of technological convergence and rapid industry shifts. Machine learning (ML) offers a more dynamic and data-driven alternative by analyzing company visuals—such as logos, product images, and branding elements—to identify similarities that go beyond predefined classifications. A recent study applies this approach to construct new industry groupings and tests them in industry momentum and reversal. The results show that ML-generated groups lead to superior performance, once again highlighting the potential of image-based classification in financial analysis.

Continue reading