The bivariate current status model

TL;DR

This paper investigates the convergence rates and limit distributions of estimators for the bivariate current status model, introducing a discrete plug-in estimator and comparing it with the MLE and SMLE through simulations.

Contribution

It introduces a new discrete plug-in estimator for the bivariate current status model and analyzes its convergence rate and distribution, providing insights into the behavior of existing estimators.

Findings

Plug-in estimator converges at rate n^{1/3} with a normal limit distribution.

SMLE likely has a n^{1/3} convergence rate with a normal limit distribution.

Simulation results suggest the plug-in estimator and SMLE have smaller variance but larger bias than the sieved MLE.

Abstract

For the univariate current status and, more generally, the interval censoring model, distribution theory has been developed for the maximum likelihood estimator (MLE) and smoothed maximum likelihood estimator (SMLE) of the unknown distribution function, see, e.g., [12], [7], [4], [5], [6], [10], [11] and [8]. For the bivariate current status and interval censoring models distribution theory of this type is still absent and even the rate at which we can expect reasonable estimators to converge is unknown. We define a purely discrete plug-in estimator of the distribution function which locally converges at rate n^{1/3} and derive its (normal) limit distribution. Unlike the MLE or SMLE, this estimator is not a proper distribution function. Since the estimator is purely discrete, it demonstrates that the n^{1/3} convergence rate is in principle possible for the MLE, but whether this actually holds for the MLE is still an open problem. If the cube root n rate holds for the MLE, this would mean that the local 1-dimensional rate of the MLE continues to hold in dimension 2, a (perhaps) somewhat surprising result. The simulation results do not seem to be in contradiction with this assumption, however. We compare the behavior of the plug-in estimator with the behavior of the MLE on a sieve and the SMLE in a simulation study. This indicates that the plug-in estimator and the SMLE have a smaller variance but a larger bias than the sieved MLE. The SMLE is conjectured to have a n^{1/3}-rate of convergence if we use bandwidths of order n^{-1/6}. We derive its (normal) limit distribution, using this assumption. Finally, we demonstrate the behavior of the MLE and SMLE for the bivariate interval censored data of [1], which have been discussed by many authors, see e.g., [18], [3], [2] and [15].