Code and results for: Sharp information bounds and global convergence of the EM algorithm for normal data under random right censoring
Year: 2026
Type: dataset
Abstract: Python code and generated results accompanying a study of the EM algorithm for the mean of a normal sample with known variance under independent random right censoring. The code verifies the lemma on the inverse Mills ratio, the identity h'(q) = (λ(a) − a)², the information–rate identity, the bounds h(p) ≤ γ(G) < p on 20,000 random censoring laws, and regenerates the numerical table and Figure 1. Run code/censoring_em_theory.py first, then code/make_figure.py. See README.txt.
Source: Mendeley Data
DOI: 10.17632/z9c5bp7y2w
Author: Berdimuradov, Mirkamol
Language: English
Subfield: Statistics and Probability
Field: Mathematics
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7225754336
