Semiparametric Bootstrap Prediction Intervals in time Series

One of the main goals of studying the time series is estimation of prediction interval based on an observed sample path of the process. In recent years, different semiparametric bootstrap methods have been proposed to find the prediction intervals without any assumption of error distribution. In semiparametric bootstrap methods, a linear process is approximated by an autoregressive process. Then the bootstrap samples are generated by resampling from the residuals. In this paper, first these sieve bootstrap methods are defined and, then, in a simulation study sieve bootstrap prediction intervals are compared with a standard Gaussian prediction interval. Finally, these methods are used to find the prediction intervals for weather data of Isfahan.