cdt15 / lingam

Python package for causal discovery based on LiNGAM.
https://sites.google.com/view/sshimizu06/lingam
MIT License
372 stars 56 forks source link

Change in method of calculating total effect using regression #137

Closed ikeuchi-screen closed 4 months ago

ikeuchi-screen commented 4 months ago

When using regression to calculate the overall effect, the coefficients were pruned using AdaptiveLasso. As a result, the distribution of the overall effect took on a skewed shape when bootstrapped. To solve this problem, we change from AdaptiveLasso to LinerRegression.

https://github.com/cdt15/lingam/blob/0cccccda96fdf7953c10c978e59bb6345c3edf36/lingam/base.py#L87

ikeuchi-screen commented 4 months ago

Changed in pull request #138