README.md

In ericstrobl/RCI: Root Causal Inference

Root Causal Inference (RCI)

RCI is an algorithm that discovers the root causes of disease in a patient-specific manner. The algorithm extracts mutually independent errors from the data assuming that the causal relations can be summarized using a linear structural equation model. RCI then outputs patient-specific Shapley values of the log-odds using logistic regression.

The academic article describing RCI in detail can be found here. The Experiments folder contains code to replicate the experimental results in the paper. Please cite the article if you use any of the code in this repository.

Installation

library(devtools)

install_local("Directory of RCI-main.zip")

library(RCI)

Example

Generate a random DAG with 10 variables and an expected neighborhood size of 2:

G = generate_DAG(p=10,en=2)

Create a dataset of 1000 samples from the DAG:

data = sample_DAG_Y(1000,G)

X = data$data[,-data$Y]; Y = data$data[,data$Y]

Run the RCI algorithm and print the Shapley values. Note that the column names correspond to the variable order in X.

out = RCI(X,Y)

print(out$scores)

ericstrobl/RCI documentation built on May 29, 2022, 10:11 p.m.