Estimation for the Half-triangle Distribution based on Generalized Adaptive Progressive Hybrid Censored Samples

Abstract

Generalized adaptive progressive hybrid censoring scheme is designed to correct the drawbacks in the adaptive progressive hybrid censoring scheme. The survival test based on the generalized adaptive progressive hybrid censoring scheme can save both the total time and cost on tests. In this paper, the estimation of the parameter and reliability of a half-triangle distribution (HaTrD) based on generalized adaptive progressive hybrid censoring scheme has been considered. Since the maximum likelihood estimators (MLE) of the parameter and reliability cannot be obtained in closed form, Newton-Raphson method have been considered. Also, we consider the approximate MLE (AMLE) using Taylor series expansion. Simulation experiments are performed to see the effectiveness of the different estimators. AMLEs of parameter and reliability perform better than its MLEs in terms of MSEs. Among AMLEs of parameter and reliability, we can observe that AMLE have overall lower MSEs. Also, a simulated data set based on the generalized adaptive progressive hybrid censoring is also analyzed for illustrative purposes.