Estimation of Right-censored SETAR-type Nonlinear Time-series Model

This paper focuses on estimating the Self-Exciting Threshold Autoregressive (SETAR) type time-series model under right-censored data. As is known, SETAR used when underlying function of relation-ship between itself ( Y t ), and its p delays $$({Y_{t - j}})_{j = 1}^p$$ violates lin-earity assumption this formed by multiple behaviors that called regime. addresses dependent problem which has a serious negative effect estimation performance. Right-censored time series cause biased coefficient estimates unqualified predictions. The main contribution solving censorship for three different techniques are kNN imputation represents techniques, Kaplan-Meier weights applied based weighted least squares, synthetic data transformation adds to modeling process manipulating dataset. Then, these solutions combined SETAR-type process. To observe behavior nonlinear estimators in practice, simulation study real example carried out. Covid-19 dataset collected China as Results prove although show satisfying performance, quality estimate technique dominates other two estimators.