DPI-package

<img src="figures/logo.png?package=DPI&version=2026.2" style="float: right" alt="logo" width="120">
The Directed Prediction Index ('DPI') is a causal discovery method for observational data designed to quantify the relative endogeneity of outcome (Y) versus predictor (X) variables in regression models. By comparing the coefficients of determination (R-squared) between the Y-as-outcome and X-as-outcome models while controlling for sufficient confounders and simulating k random covariates, it can quantify relative endogeneity, providing a necessary but insufficient condition for causal direction from a less endogenous variable (X) to a more endogenous variable (Y). Methodological details are provided at <a href="https://psychbruce.github.io/DPI/">https://psychbruce.github.io/DPI/</a>. This package also includes functions for data simulation and network analysis (correlation, partial correlation, and Bayesian Networks).

internal

The Directed Prediction Index ('DPI') is a causal discovery method
for observational data designed to quantify the relative endogeneity
of outcome (Y) versus predictor (X) variables in regression models.
By comparing the coefficients of determination (R-squared)
between the Y-as-outcome and X-as-outcome models
while controlling for sufficient confounders and
simulating k random covariates, it can quantify relative endogeneity,
providing a necessary but insufficient condition for causal direction
from a less endogenous variable (X) to a more endogenous variable (Y).
Methodological details are provided at
<https://psychbruce.github.io/DPI/>.
This package also includes functions for data simulation and network
analysis (correlation, partial correlation, and Bayesian Networks).

Han-Wu-Shuang Bao

The Directed Prediction Index for Causal Direction Inference
from Observational Data

DPI-package function

Maintainer: Han Wu Shuang Bao <a href="/link/baohws%40foxmail.com?package=DPI&version=2026.2" data-mini-rdoc="DPI::baohws@foxmail.com">baohws@foxmail.com</a> (<a href="https://orcid.org/0000-0003-3043-710X">ORCID</a>)

Author

<img src='figures/logo.png' style='float: right' alt='logo' width='120' />
The Directed Prediction Index ('DPI') is a causal discovery method for observational data designed to quantify the relative endogeneity of outcome (Y) versus predictor (X) variables in regression models. By comparing the coefficients of determination (R-squared) between the Y-as-outcome and X-as-outcome models while controlling for sufficient confounders and simulating k random covariates, it can quantify relative endogeneity, providing a necessary but insufficient condition for causal direction from a less endogenous variable (X) to a more endogenous variable (Y). Methodological details are provided at <a href='https://psychbruce.github.io/DPI/'>https://psychbruce.github.io/DPI/</a>. This package also includes functions for data simulation and network analysis (correlation, partial correlation, and Bayesian Networks).

DPI: The Directed Prediction Index for Causal Direction Inference from Observational Data — DPI-package

Maintainer: Han Wu Shuang Bao <a href='mailto:baohws@foxmail.com'>baohws@foxmail.com</a> (<a href='https://orcid.org/0000-0003-3043-710X'>ORCID</a>)

DPI-package: DPI: The Directed Prediction Index for Causal Direction Inference from Observational Data

Description

Arguments

Author

See Also