Industry-Neutral Factor Return Regression and Significance Tests

Summary

The document presents a workflow for evaluating a stock factor with Chinese A-share data. It builds daily stock returns and industry returns, represents industry membership with indicator variables, and subtracts each stock’s industry return to form an industry-adjusted outcome. For each date, it regresses that outcome on a selected factor without an intercept, recording the factor’s estimated return, t-statistic, and R-squared.

It then summarizes the daily absolute t-statistics with their mean, the share reaching a stated threshold, and an aggregate statistic based on their dispersion and count. The example uses a volatility-related factor and describes preprocessing that includes winsorization and date-wise standardization. The document provides implementation code and a workflow rather than reported empirical findings. It does not explain whether the regression assumptions hold, how missing or failed daily fits affect the summary, or whether the aggregate significance calculation is statistically appropriate; its results therefore require independent validation.

Key ideas

• Daily factor returns are estimated by regressing industry-adjusted stock returns on factor exposures without an intercept.
• Industry returns are approximated using stock industry membership and corresponding industry index returns.
• The workflow records daily factor estimates, t-statistics, and R-squared values.
• The final summary uses absolute daily t-statistics, including their mean and the fraction above a threshold.
• Preprocessing includes winsorization and cross-sectional standardization by date.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.