
Causal factor investing takes a scientific approach to identifying and investing in specific factors that drive returns, rather than relying on traditional asset class or sector-based investing. This approach is based on the idea that certain factors, such as value, momentum, and size, have consistently outperformed the market over time.
By using statistical models and machine learning algorithms, causal factor investors can identify the underlying drivers of returns and make more informed investment decisions. For example, research has shown that a portfolio of small-cap stocks has historically outperformed a portfolio of large-cap stocks.
Investors who use a causal factor approach can create a diversified portfolio that captures the benefits of different factors, such as value and momentum. This can help to reduce risk and increase potential returns.
Check this out: What Are the Factors of 56?
Scientific Approach
The current state of factor investing is stuck in a phenomenological stage, meaning we don't have a clear understanding of the underlying causal mechanisms.
Virtually all journal articles in the factor investing literature make associational claims, rather than causal claims. This is because authors don't identify the causal graph consistent with the observed phenomenon.
Authors often justify their chosen model specification in terms of correlations, but this is not enough to establish causality. They also fail to propose experiments for falsifying causal mechanisms.
Absent a causal theory, their findings are likely false, due to rampant backtest overfitting and incorrect specification choices. This is a major problem in the field of factor investing.
The article notes that there are two types of spurious claims in the factor investing literature: type-A and type-B. Both types prevent factor investing from advancing beyond its current stage.
To transform factor investing into a truly scientific discipline, we need to adopt a more rigorous approach. This involves identifying the causal graph consistent with the observed phenomenon and proposing experiments to test causal mechanisms.
See what others are reading: B Riley Preferred Stock
Featured Images: pexels.com


