Nonparametric estimation of a distribution function from doubly truncated data under dependence

The NPMLE of a distribution function from doubly truncated data was introduced in the seminal paper Efron and Petrosian (J Am Stat Assoc 94:824–834, 1999). consistency depends however on assumption independent truncation. In this work we introduce an extension Efron–Petrosian when variable interest truncation variables may be dependent. proposed estimator is constructed basis copula which represents dependence structure between variables. Two different iterative algorithms to compute practice are introduced, their performance explored through intensive Monte Carlo simulation study. We illustrate use estimators two real examples.