Forecasting time series with sieve bootstrap
نویسندگان
چکیده
منابع مشابه
Sieve Bootstrap for Time Series Sieve Bootstrap for Time Series
We study a bootstrap method which is based on the method of sieves. A linear process is approximated by a sequence of autoregressive processes of order p = pn, where pn ! 1 ; p n = on as the sample size n ! 1. F or given data, we t h e n estimate such a n A R pn model and generate a bootstrap sample by resampling from the residuals. This sieve bootstrap enjoys a nice nonparametric property. We ...
متن کاملForecasting nonlinear time series with neural network sieve bootstrap
A new method to construct nonparametric prediction intervals for nonlinear time series data is proposed. Within the framework of the recently developed sieve bootstrap, the new approach employs neural network models to approximate the original nonlinear process. The method is flexible and easy to implement as a standard residual bootstrap scheme while retaining the advantage of being a nonparam...
متن کاملThe bootstrap methodology in time series forecasting
The bootstrap methodology, initially proposed in independent situations, has revealed inefficient in the context of dependent data. Here, the estimation of population characteristics is more complex. This is what happens in the context of time series. There has been a great development in the area of resampling methods for dependent data. A revision of different approaches of this methodology f...
متن کاملAn Invariance Principle for Sieve Bootstrap in Time Series
This paper establishes an invariance principle applicable for the asymptotic analysis of sieve bootstrap in time series+ The sieve bootstrap is based on the approximation of a linear process by a finite autoregressive process of order increasing with the sample size, and resampling from the approximated autoregression+ In this context, we prove an invariance principle for the bootstrap samples ...
متن کاملNeural network sieve bootstrap prediction intervals for hydrological time series
When analyzing time series data, the estimation of forecast intervals, based on an observed sample path of the process, is a key issue. If the process is linear and the distribution of the error process is known, the methodology is well developed but, for departures from the true underlying distribution, the prediction intervals perform poorly. In this latter case several distribution free alte...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Journal of Statistical Planning and Inference
سال: 2002
ISSN: 0378-3758
DOI: 10.1016/s0378-3758(01)00092-1