Improving Equity Model Fit with Factors and Out-of-Sample Validation
Summary
The document considers why an intraday linear equity-return model may fit some stocks better than others and whether adding beta, liquidity, or sector returns would make explained-variance statistics more consistent. It outlines factor-based approaches, including market exposure, size and value extensions, sector or industry comparisons, and broader factor models. Principal component analysis or fundamental research can help identify candidate factors, while regional exposures may matter for globally traded portfolios.
The responses caution against judging predictive models primarily by in-sample R-squared and significance. Cross-validation and out-of-sample prediction are more useful for assessing whether apparent relationships generalize. They also recommend comparing the characteristics of well-fit and poorly fit stocks: uneven results may reflect a sector effect or other shared exposure that should be modeled or controlled. The discussion offers diagnostic directions rather than evidence that adding particular factors will improve performance; factor choices require investigation and validation on the relevant data.
Key ideas
- Market, size, value, sector, and other exposures can serve as candidate explanatory factors for equity returns.
- Compare the characteristics of stocks with strong and weak model fit to identify hidden group effects.
- Sector controls or neutrality can help distinguish industry effects from other signals.
- Use cross-validation and out-of-sample predictions to assess predictive value instead of relying on in-sample fit alone.
- Additional factors can increase model complexity, so their usefulness must be validated.
Tags
Full text
# How to improve the consistency of explained variance statistics in a linear equity model? # How to improve the consistency of explained variance statistics in a linear equity model? I have an intraday equity returns linear model that, overall, shows good values in terms of $R^2$, p-value and other explained variance statistics. Around 70% of the stocks show consistent R-squared (in the range of 5% explained) and significance parameters, but the remaining 30% have degraded $R^2$ and p-values. Would it help to include beta, liquidity and sector returns in the model to have a more consistent $R^2$ across all stocks? ## Answer by chrisaycock (score 5, accepted) https://quant.stackexchange.com/a/2686 Since you mention beta, I assume you're familiar with the capital asset pricing model (CAPM). The concept is that an asset's expected returns are linearly correlated with the market's returns. Of course, there are other ways "normalize" returns, as you put it. We can extend CAPM with Fama-French, which adds market cap and relative value to the equation. Within the realm of statistical arbitrage, sector-neutrality is very common. Stocks are compared with their sector or industry peers. An added dimension is region or country affiliation, if you were to trade globally. Then there's arbitrage pricing theory (APT), which defines an asset's price in terms of numerous possible factors. I wouldn't call it "state of the art", but it takes a more realistic view of fair value. You'll need to define your own factors, which you can cull through either fundamental analysis or principal component analysis. You can also buy commercial risk models from numerous vendors. In this case, you may find yourself balancing stocks within the portfolio according to their exposure to things like interest rates or the price of oil. ## Answer by Patrick Burns (score 3) https://quant.stackexchange.com/a/2688 When trying to predict returns, I think you should never look at in-sample statistics like R-squared. Only look at out-of-sample prediction results. Cross validation is a useful tool in at least the initial phase of modelling. In addition to over-enthusiasm, in-sample statistics easily lead to overfitting: http://www.portfolioprobe.com/2011/03/28/the-devil-of-overfitting/ ## Answer by Tal Fishman (score 1) https://quant.stackexchange.com/a/2704 I suggest you look for common characteristics amongst the 70% of stocks with decent $R^2$ and the 30% with a degraded in-sample fit. If you find that all the stocks that fit your model best were in one industry and all those that didn't fit were in another industry, then it may be that your model is actually picking up an industry effect. Controlling for sector/industry explicitly as part of your model, whether by imposing sector neutrality in the weights for a hypothetical backtest portfolio or by including it as an additional factor in the linear model, may reveal what is really going on. In short, these models can be quite complicated, and there is no shortcut around doing the dirty work and looking in depth at a lot of specific cases to learn more about what your model is really doing.
Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.