Edgeworth Correction by Bootstrap in Autoregressions
Bose, Arup
Ann. Statist., Tome 16 (1988) no. 1, p. 1709-1722 / Harvested from Project Euclid
Text on Project Euclid
PDF
Alt PDF
We prove that the distribution of least-squares estimates in autoregressions can be bootstrapped with accuracy $o(n^{-1/2})$ a.s., thereby improving the normal approximation error of $O(n^{-1/2})$.
Publié le : 1988-12-14
Classification:  Autoregressions,  Cramer's condition,  Edgeworth expansion,  bootstrap,  62F12,  62M10,  60F99,  60G10 
@article{1176351063,
     author = {Bose, Arup},
     title = {Edgeworth Correction by Bootstrap in Autoregressions},
     journal = {Ann. Statist.},
     volume = {16},
     number = {1},
     year = {1988},
     pages = { 1709-1722},
     language = {en},
     url = {http://dml.mathdoc.fr/item/1176351063}
}

Bose, Arup. Edgeworth Correction by Bootstrap in Autoregressions. Ann. Statist., Tome 16 (1988) no. 1, pp.  1709-1722. http://gdmltest.u-ga.fr/item/1176351063/