This factor is designed to capture the market capitalization scale effect of stocks. The market capitalization distribution of stocks in the A-share market has obvious thick-tail characteristics, that is, a few large-cap stocks occupy most of the market share, while a large number of small-cap stocks are distributed in the tail of the distribution. Directly using market capitalization as a factor may cause factor analysis to be greatly affected by extreme values ​​and it is difficult to meet the model's assumption of normal distribution. By taking the logarithm of market capitalization, the dynamic range of market capitalization data can be compressed, making its distribution closer to normal distribution, reducing the impact of extreme values, and thus improving the effectiveness and robustness of factor analysis. In addition, the logarithm-transformed market capitalization factor usually shows a stronger linear relationship in the quantitative model and is easier to be learned and utilized by the model.