Asymptotic normality of conditional density estimation under truncated, censored and dependent data

Abstract

In this paper, we focus on the left-truncated and right-censored model, and construct the local linear and Nadaraya-Watson type estimators of the conditional density. Under suitable conditions, we establish the asymptotic normality of the proposed estimators when the observations are assumed to be a stationary α-mixing sequence. Finite sample behavior of the estimators is investigated via simulations too.

Keywords:

• Asymptotic normality
• local linear and Nadaraya-Watson type estimators
• conditional density
• truncated and censored data
• α-mixing

AMS 2000 subject classifications:

• 62N01
• 62E20

Additional information

Funding

This research was supported by the National Social Science Foundation of China (17BTJ032).