An alternative block bootstrap in time series with weak dependence
Abstract views: 3 / PDF downloads: 2
Keywords:Bootstrap, $\alpha$-mixing, time series, weak dependence, m-dependent
When the observations are dependent, like in time series, Kunsch introduced the bootstrap with blocks forming by a fix number of consecutive observations. Different versions of block bootstrap has been formulated. In this paper we have proposed a bootstrap estimation with blocks formed from recalculated values of a statistic. We call it bootstrap with re-blocks. We have shown that this bootstrap works in time series strictly stationary $\alpha$-mixing or $m $-dependent under some conditions. We have done
simulations to compare the bootstrap with re-blocks with other block bootstrap methods.
How to Cite
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.