Partial Least Squares Path Modeling

Introduction

This is a python implementation of Partial Least Squares Path Modeling for metric data, based on plspm of R version.

Path analysis is also known as Structural Equation Modeling(SEM). It can model causality, on which statistical model would not work.

Further, Judea Pearl's causal inference is at the cutting edge of causal analysis, which derived from bayesian network and has attracted many people's attention, especially those who work on social science. Pearl's book "The Book of Why" has an introduction to the new paradigm.

Code

plspm.py

Core program

test_plspm.py

Test Program

utils.py

Utils

Reference

1. http://www.gastonsanchez.com/PLS_Path_Modeling_with_R.pdf
2. http://causality.cs.ucla.edu/blog/