Multi-Innovation Newton Recursive Methods of an Exponential Autoregressive Time Series Model Based on the Penalty Term
摘要
This paper focuses on the parameter estimation of exponential autoregressive (ExpAR) models. Employing the Newton search and the penalty term, a Newton recursive algorithm is proposed for estimating the parameters of the ExpAR model. To achieve higher accuracy performance under white noise interference, a multi-innovation Newton recursive algorithm is proposed to make more use of the useful data. The proposed algorithm is tested by simulating experiments to see if it works well.