Proves √n-excess risk bounds, consistency, convergence rates, asymptotic normality, and a sufficient condition on missing probability and cluster separation for k-means under MCAR missing data.
Title resolution pending
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
stat.ML 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
MNAR-k-means constrains imputation values in k-means for magnitude-decaying MNAR missingness and establishes statistical consistency of the resulting cluster centers to those of fully observed data.
citing papers explorer
-
Statistical Properties of $k$-means Clustering for Data Missing Completely at Random
Proves √n-excess risk bounds, consistency, convergence rates, asymptotic normality, and a sufficient condition on missing probability and cluster separation for k-means under MCAR missing data.
-
MNAR-$k$-means: A $k$-means Clustering for Data Missing Not at Random with Magnitude-Decaying Probability
MNAR-k-means constrains imputation values in k-means for magnitude-decaying MNAR missingness and establishes statistical consistency of the resulting cluster centers to those of fully observed data.