Advanced

Two Sides of the Same Coin: Bootstrapping the Restricted vs. Unrestricted Model

Mantalos, Panagiotis LU (2005) In Journal of Modern Applied Statistical Methods 4(1). p.35-42
Abstract
The properties of the bootstrap test for restrictions are studied in two versions: 1) bootstrapping under the null hypothesis, restricted, and 2) bootstrapping under the alternative hypothesis, unrestricted. This article demonstrates the equivalence of these two methods, and illustrates the small sample properties of the Wald test for testing Granger-Causality in a stable stationary VAR system by Monte Carlo methods. The analysis regarding the size of the test reveals that, as expected, both bootstrap tests have actual sizes that lie close to the nominal size. Regarding the power of the test, the Wald and bootstrap tests share the same power as the use of the Size-Power Curves on a correct size-adjusted basis.
Please use this url to cite or link to this publication:
author
organization
publishing date
type
Contribution to journal
publication status
published
subject
keywords
VAR system, words: Bootstrap, Granger-Causality, Wald test
in
Journal of Modern Applied Statistical Methods
volume
4
issue
1
pages
35 - 42
publisher
JMASM
external identifiers
  • scopus:22944440970
ISSN
1538-9472
language
English
LU publication?
yes
id
b105c094-a165-42ba-8bf5-f6a0797ebb79 (old id 1387099)
alternative location
http://tbf.coe.wayne.edu/jmasm/vol4_no1.pdf
date added to LUP
2009-04-20 12:27:24
date last changed
2017-01-01 06:40:46
@article{b105c094-a165-42ba-8bf5-f6a0797ebb79,
  abstract     = {The properties of the bootstrap test for restrictions are studied in two versions: 1) bootstrapping under the null hypothesis, restricted, and 2) bootstrapping under the alternative hypothesis, unrestricted. This article demonstrates the equivalence of these two methods, and illustrates the small sample properties of the Wald test for testing Granger-Causality in a stable stationary VAR system by Monte Carlo methods. The analysis regarding the size of the test reveals that, as expected, both bootstrap tests have actual sizes that lie close to the nominal size. Regarding the power of the test, the Wald and bootstrap tests share the same power as the use of the Size-Power Curves on a correct size-adjusted basis.},
  author       = {Mantalos, Panagiotis},
  issn         = {1538-9472},
  keyword      = {VAR system,words: Bootstrap,Granger-Causality,Wald test},
  language     = {eng},
  number       = {1},
  pages        = {35--42},
  publisher    = {JMASM},
  series       = {Journal of Modern Applied Statistical Methods},
  title        = {Two Sides of the Same Coin: Bootstrapping the Restricted vs. Unrestricted Model},
  volume       = {4},
  year         = {2005},
}