Showing 1 - 10 of 18 results

Sort By
  • Relevance
  • Newest
  • Oldest

Working Paper

Understanding The Virtue of Complexity

We respond to recent academic challenges to aspects of the “virtue of complexity” described in our prior research. We provide detailed discussions of how complex models learn in small samples, the roles of “nominal” and “effective” complexity, the unique effects of implicit regularization, and the importance of limits to learning. We then present new empirical and theoretical analyses that expand on KMZ. Finally, we introduce and demonstrate the virtue of ensemble complexity.

Alternative Thinking

Can Machines Build Better Stock Portfolios?

In the second issue of our 2024 Alternative Thinking series, we showed that machine learning techniques can be used to help improve market timing strategies. In this issue, we extend these concepts to constructing stock selection strategies following a similar framework. Our results indicate more complex models utilizing machine learning techniques yield performance improvements relative to a simple, linear approach in the range of 50-100%, suggesting that machine learning can help to build better stock selection portfolios.

Journal Article

CIO Perspectives: An Interview with Cliff Asness

In a wide ranging interview, AQR managing principal Cliff Asness discusses many aspects of AQR’s investment philosophy and approach from the perspective of a CIO – how we adapt our process to changing market conditions, how we think about adding innovative technology such as machine learning to our process, and more.

Journal Article

Business News and Business Cycles

We propose an approach to measuring the state of the economy via textual analysis of business news. From the full text of 800,000 Wall Street Journal articles for 1984 to 2017, we estimate a topic model that summarizes business news into interpretable topical themes and quantifies the proportion of news attention allocated to each theme over time. News attention closely tracks a wide range of economic activities and can forecast aggregate stock market returns.

Alternative Thinking

Can Machines Time Markets? The Virtue of Complexity in Return Prediction

Common wisdom has suggested that small, simple models are best suited for market timing applications, given finance’s “small data” constraint and naturally low predictability. However, we show that complex models better identify true nonlinear relationships and therefore produce better market timing strategy performance. We validate this "virtue of complexity" result in three practical market timing applications.

Journal Article

The Virtue of Complexity in Return Prediction

Contrary to conventional wisdom, we theoretically prove that simple models severely understate return predictability compared to “complex” models in which the number of parameters exceeds the number of observations. We empirically document the virtue of complexity in U.S. equity market return prediction. Our findings establish the rationale for modeling expected returns through machine learning.

Working Paper

How Global is Predictability?

We show that asset pricing has a strong global component in the sense that a common global model has stronger predictability of stock returns than local models estimated in each country – even when the global model is estimated without the use of local data. Nevertheless, asset pricing has a small local component – in order to detect it, we develop a refined transfer learning model that gains power and precision by building off the global component.

Working Paper

Financial Machine Learning

In this survey the nascent literature on machine learning in financial markets, we highlight the best examples of what this line of research has to offer and recommend promising directions for future research.

Working Paper

Machine Learning and the Implementable Efficient Frontier

We propose that investment strategies should be evaluated based on their net-of-trading-cost return for each level of risk, which we term the "implementable efficient frontier." While numerous studies use machine learning return forecasts to generate portfolios, their agnosticism toward trading costs leads to excessive reliance on fleeting small-scale characteristics, resulting in poor net returns. We develop a framework that produces a superior frontier by integrating trading-cost-aware portfolio optimization with machine learning

Alternative Thinking

Can Machine Learning Help Manage Climate Risks?

Some investors have incorporated carbon emissions into investment selection as their primary approach to preparing their portfolios for a future regime shift to a lower-carbon economy. However, carbon emissions can be a narrow measure of overall climate risk. To complement this approach, we explore how machine learning techniques may be able create a broad climate hedging portfolio based on stocks’ sensitivity to climate news.