Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
1151703 | Statistics & Probability Letters | 2013 | 10 Pages |
Abstract
Consider a nearly nonstationary AR(1) model, Xt=θnXtâ1+ut, where θn=1âγ/n, γ is a fixed constant, and the innovations are in the domain of attraction of the normal law with possibly infinite variance. As for the least squares estimator θËn of θn, we propose to use a residual-based m-out-of-n bootstrap procedure to approximate the distribution of θËnâθn, and its asymptotic validity is proved.
Related Topics
Physical Sciences and Engineering
Mathematics
Statistics and Probability
Authors
Ke-Ang Fu, Yuechao Li, Andrew Cheuk-Yin Ng,