Assumption-Lean and Data-Adaptive Post-Prediction Inference.

POP-Inf

This repository hosts the R package that implements the POP-Inf method described in the paper: Assumption-lean and data-adaptive post-prediction inference.

POP-Inf provides valid and powerful inference based on ML predictions for parameters defined through estimation equations.

Installation

# install.packages("devtools")
devtools::install_github("qlu-lab/POPInf")

Useful examples

Here are examples of POP-Inf for M-estimation tasks including: mean estimation, linear regression, logistic regression, and Poisson regrssion. The main function is pop_M(), where the argument method indicates which task to do.

# Load the package
library(POPInf)

# Load the simulated data
set.seed(999)
data <- sim_data()
X_lab = data$X_lab ## Covariates in the labeled data
X_unlab = data$X_unlab ## Covariates in the unlabeled data
Y_lab = data$Y_lab ## Observed outcome in the labeled data
Yhat_lab = data$Yhat_lab ## Predicted outcome in the labeled data
Yhat_unlab = data$Yhat_unlab ## Predicted outcome in the unlabeled data

Mean estimation

# Run POP-Inf mean estimation
fit_mean <- pop_M(Y_lab = Y_lab, Yhat_lab = Yhat_lab, Yhat_unlab = Yhat_unlab,
                  alpha = 0.05, method = "mean")

print(fit_mean)

#   Estimate  Std.Error Lower.CI Upper.CI       P.value    Weight
# 1 1.623601 0.05514429  1.51552 1.731682 1.557956e-190 0.9226747

Linear regression

# Run POP-Inf linear regression
fit_ols <- pop_M(X_lab = X_lab, X_unlab = X_unlab,
           Y_lab = Y_lab, Yhat_lab = Yhat_lab, Yhat_unlab = Yhat_unlab,
           alpha = 0.05, method = "ols")

print(fit_ols)

#     Estimate  Std.Error  Lower.CI Upper.CI       P.value    Weight
#    1.6181357 0.05351775 1.5132429 1.723029 8.093485e-201 0.8811378
# X1 0.8716172 0.07335443 0.7278452 1.015389  1.463365e-32 1.0000000

Logistic regression

# Load the simulated data
set.seed(999)
data <- sim_data(binary = T)
X_lab = data$X_lab
X_unlab = data$X_unlab
Y_lab = data$Y_lab
Yhat_lab = data$Yhat_lab
Yhat_unlab = data$Yhat_unlab

# Run POP-Inf logistic regression
fit_logistic <- pop_M(X_lab = X_lab, X_unlab = X_unlab,
                      Y_lab = Y_lab, Yhat_lab = Yhat_lab, Yhat_unlab = Yhat_unlab,
                      alpha = 0.05, method = "logistic")

print(fit_logistic)

#      Estimate  Std.Error   Lower.CI   Upper.CI      P.value    Weight
#    -0.1355928 0.08443198 -0.3010764 0.02989085 1.082868e-01 0.4218688
# X1  0.5876862 0.08938035  0.4125039 0.76286842 4.861518e-11 0.5340878

Poisson regression

# Load the simulated data
set.seed(999)
data <- sim_data()
X_lab = data$X_lab
X_unlab = data$X_unlab
Y_lab = round(data$Y_lab - min(data$Y_lab))
Yhat_lab = round(data$Yhat_lab - min(data$Yhat_lab))
Yhat_unlab = round(data$Yhat_unlab - min(Yhat_unlab))

# Run POP-Inf Poisson regression
fit_poisson <- pop_M(X_lab = X_lab, X_unlab = X_unlab,
                     Y_lab = Y_lab, Yhat_lab = Yhat_lab, Yhat_unlab = Yhat_unlab,
                     alpha = 0.05, method = "poisson")

print(fit_poisson)

#     Estimate  Std.Error  Lower.CI  Upper.CI      P.value    Weight
#    1.2227700 0.01730779 1.1888473 1.2566926 0.000000e+00 0.8460699
# X1 0.2568325 0.02437762 0.2090532 0.3046118 5.921326e-26 0.9171068

Analysis script

We provide the script for analysis in the POP-Inf paper here.

Contact

Please submit an issue or contact Jiacheng ([email protected]) or Xinran ([email protected]) for questions.

Reference

Assumption-lean and Data-adaptive Post-Prediction Inference

Valid inference for machine learning-assisted GWAS

"POP" familial links

• POP-TOOLS (POst-Prediction TOOLS) is a toolkit for conducting valid and powerful machine learning (ML)-assisted genetic association studies. It currently implements
  • POP-GWAS, where statistical and computational methods are optimized for GWAS applications.