pseudoSCR

A Pseudo-Value Approach to Causal Deep Learning of Semi-Competing Risks

Overview

This project proposes a deep learning approach for estimating the causal effect of treatment on non-fatal outcomes in the presence of dependent censoring and complex covariate relationships. Our three-stage approach involves:

1. Estimating the non-fatal event’s marginal survival function using an Archimedean copula representation
2. Constructing jackknife pseudo-values that estimate pseudo-survival probabilities for the non-fatal event at fixed time points
3. Fitting a deep neural network (S-learner) to estimate survival average causal effect of treatment

• GitHub: https://github.com/salernos/pseudoSCR
• Paper: https://link.springer.com/article/10.1007/s40065-025-00501-7