Interpreting Insignificant Fama–French Factor Betas

Summary

The document describes a thesis analysis using randomly selected stocks grouped into portfolios and asks why estimated Fama–French three-factor betas, especially SMB and HML, have p-values above 0.05. The proposed workflow constructs SMB and HML portfolios using the stocks in the sample portfolios, estimates each stock’s factor loadings by regression, and plans to average those estimates over a period to calculate expected portfolio returns. The selected observation period runs from 2018 to 2023.

No response or diagnostic analysis is included, so the document does not identify a cause for the high p-values. It raises a useful empirical question about statistical significance in factor regressions, but gives no details on return frequency, sample size, standard errors, factor construction, or regression specification. Those omissions limit conclusions: the reported p-values alone do not show whether the issue lies in weak factor exposure, noisy estimates, overlapping portfolio construction, or model setup.

Key ideas

• The analysis estimates stock-level loadings on the market, size, and value factors using regression.
• The author reports p-values above 0.05, particularly for the size and value factors.
• The study covers 2018 through 2023 and intends to average estimated loadings for portfolio return expectations.
• The document provides no answer, so the cause of the weak statistical significance remains undetermined.

Tags

# Issue in Fama-French 3 factors Model


# Issue in Fama-French 3 factors Model












I hope you are doing well. As part of my thesis, I built 12 portfolios of 200 randomly selected stocks. I now want to calculate the Bs of the 3 Fama French factors for each stock, so that I can average them for a time T and calculate the expected return on my portfolio at time T. I have therefore constructed the HML and SMB portfolios on the basis of the stocks in my 12 portfolios. The problem is that when I run my regression to calculate the Beta of the factors (B1, B2, B3) for a stock, the p-values of the various factors (in particular SMB and HML) are above 0.05. Could you explain this? Do you have a possible explanation? I'm doing this exercise on the period from 2018 to 2023.

Thank you very much in advance for your help.

Regards.

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.