High-dimensional censored MIDAS logistic regression for corporate survival forecasting

The paper introduces a high‑dimensional censored MIDAS logistic regression for forecasting corporate distress, addressing right censoring, high‑dimensional mixed‑frequency predictors, and mixed‑frequency data. It uses inverse probability weighting for censoring, a sparse‑group penalty for estimation, and develops a de‑sparsified estimator with asymptotic theory that accounts for censoring and heavy tails. The method is validated via Monte Carlo simulations and applied to Chinese‑listed firms, with implementation available in the R package Survivalml.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Statistics ML.