Non-Binary Nature of Theoretical Paths and Theoretical Differentiation

A brief introduction to multi-factor strategy development: Multi-factor strategy development is a type of quantitative stock selection strategy. It involves identifying filtering conditions with positive value for stock selection from fundamental, technical, or other perspectives. By stacking multiple filtering conditions, the excess return potential of each condition is compounded. In the industry, these filtering conditions are referred to as factors, and this investment strategy development path is known as the “multi-factor stock selection strategy.”

During the development of my multi-factor investment strategies in the past, I found that some factors, when used independently, did not exhibit a significant excess return value. However, in specific strategy directions, they performed exceptionally well. For example, when a factor combination leaned toward a momentum strategy, incorporating a filter for the top 20% quantile of stock price increases over the past six months could improve backtest performance. However, applying the same filter in a value strategy would degrade backtest performance. Similarly, using the lowest 20% quantile of the price-to-book (P/B) ratio as a screening factor could significantly benefit a multi-factor combination that leans toward value investment strategies. Still, it would have a negative impact when applied to a momentum strategy combination.

It must be acknowledged that both value investment and momentum strategies have long-term application value in both theory and practice. However, the filtering factors that should be applied based on their respective theoretical underpinnings differ or even contradict each other. This demonstrates that two completely opposing factor filters can provide positive value under different theoretical paths.

Based on the above experience in investment strategy development, it is clear that theoretical paths are not binary oppositions; two conflicting theoretical systems may both have positive value. Some operations or perspectives that are valuable in theoretical system A may have no value in system B. This means that when observing and understanding theoretical differentiation, observers and commentators should not make simplistic binary judgments (e.g., assuming that because path A is correct, path B is incorrect). Moreover, behaviors or perspectives should not be entirely dismissed just because they lack value in a particular path. This also highlights that the adaptability of a theory is more practically significant than its truthfulness.