Abstract:The asymptotically efficient online learning problem is investigated for stochastic censored regression models, which arise from various fields of learning and statistics but up to now still lacks comprehensive theoretical studies on the efficiency of the learning algorithms. For this, we propose a two-step online algorithm, where the first step focuses on achieving algorithm convergence, and the second step is dedicated to improving the estimation performance. Under a general excitation condition on the data, we show that our algorithm is strongly consistent and asymptotically normal by employing the stochastic Lyapunov function method and limit theories for martingales. Moreover, we show that the covariances of the estimates can achieve the Cramer-Rao (C-R) bound asymptotically, indicating that the performance of the proposed algorithm is the best possible that one can expect in general. Unlike most of the existing works, our results are obtained without resorting to the traditionally used but stringent conditions such as independent and identically distributed (i.i.d) assumption on the data, and thus our results do not exclude applications to stochastic dynamical systems with feedback. A numerical example is also provided to illustrate the superiority of the proposed online algorithm over the existing related ones in the literature.

ASYMPTOTICALLY EFFICIENT ESTIMATION IN CENSORED AND TRUNCATED REGRESSION-MODELS

A Missing Information Principle and M-Estimators in Regression-Analysis with Censored and Truncated Data

ASYMPTOTIC THEORY OF A BIAS-CORRECTED LEAST-SQUARES ESTIMATOR IN TRUNCATED REGRESSION

Asymptotically Efficient Online Learning for Censored Regression Models Under Non-I.I.D Data

Rank Regression Methods for Left-Truncated and Right-Censored Data

Recursive Least Squares for Censored Regression

Estimating a Distribution Function with Truncated and Censored Data

Censored Regression with Noisy Input

Rank Estimation in Partial Linear Model with Censored Data

Asymptotic Normality of an M-estimator of regression function for truncated-censored data under alpha-mixing condition

Single-index Regression models with right-censored responses

Median Regression of Truncated and Censored Data

Regression Analysis of Dependent Current Status Data with Left-Truncation Under Linear Transformation Model

Asymptotics for censored regression quantiles

A new approach to regression analysis of censored competing-risks data

An Informative Subset-Based Estimator for Censored Quantile Regression

Asymptotic Normality of a Class of Adaptive Statistics with Applications to Synthetic Data Methods for Censored Regression

Right-censored models by the expectile method

Joint Analysis of Longitudinal Data with Informative Right Censoring

Efficient estimation for left-truncated competing risks regression for case-cohort studies

Median Regression for Left and Right Censored Data