Bootstrap prediction intervals for autoregressions
Pages 229-239
Abstract
The bootstrap technique is applied to obtain interval forecasts for an autoregressive time series. The relevant features of the proposed method are: (i) it is distribution-free, and (ii) it explicitly takes into account that order and parameters of the model are estimated from the data.
Keywords: Autoregressive models
, Bootstrap
, Prediction intervals
FULL TEXT LINK
http://dx.doi.org/10.1016/0169-2070(90)90008-Y
ONLINE SUPPLEMENTS
