id: 02370634
dt: j
an: 2006f.03873
au: Jeong, Jinook
ti: $R^2$-based bootstrap tests for nonnested hypotheses in regression models.
so: InterStat, No. 8, 20 p. (2006).
py: 2006
pu: ,
la: EN
cc: K85
ut: nonnested regression models; bootstrap; comparison of $R^2$ tests; Monte
Carlo simulation; Cox test; $J$-test
ci:
li:
ab: This paper utilizes the bootstrap to construct tests using $R^2$ for
nonnested regression models. The bootstrap enables us to compute the
statistical significance of the differences in $R^2$’s and to
formally test about nonnested regression models. Bootstrapped $R^2$
tests that this paper proposes are expected to show better finite
sample properties since they do not have such cumulated errors in the
computation process. Moreover, bootstrapped $R^2$ tests will remove the
possibility of inconsistent test results that the previous tests suffer
from. Because bootstrapped $R^2$ tests only evaluate if a model has a
significantly higher explanatory power than the other model, there is
no possibility for inconsistent results. This study presents Monte
Carlo simulation results to compare the finite sample properties of the
proposed tests with the previous tests such as Cox test and $J$-test.
(orig.)
rv: