Local Linear Estimator of the Conditional Hazard Function For Index Model in Case of Missing at Random Data

The estimation of hazard function becomes an important tool in statistics. Also, the single-index model is an essential method that has much application in a lot of domains but the use of this model in nonparametric estimation is very limited, and only a few theoretical results are available in the literature. In this paper, a kernel estimator of the conditional hazard function is proposed for the index model using the local linear model. This estimator is created under some regularity conditions and in the case of missing data at random. Then, the almost complete convergence is established with rate for the proposed estimator. The results prove that it has good asymptotic properties.