An Omnibus Test for Departures from Constant Mean
Barry, Daniel ; Hartigan, J. A.
Ann. Statist., Tome 18 (1990) no. 1, p. 1340-1357 / Harvested from Project Euclid
Observations $y_i$ are made at points $x_i$ according to the model $y_i = F(x_i) + e_i$, where the $e_i$ are independent normals with constant variance. In order to decide whether or not $F(x)$ is constant, a likelihood ratio test is constructed, comparing $F(x) \equiv \mu$ with $F(x) = \mu + Z(x)$, where $Z(x)$ is a Brownian motion. The ratio of error variance to Brownian motion variance is chosen to maximize the likelihood, and the resulting maximum likelihood statistic $B$ is used to test departures from constant mean. Its asymptotic distribution is derived and its finite sample size behavior is compared with five other tests. The $B$-statistic is comparable or superior to each of the tests on the five alternatives considered.
Publié le : 1990-09-14
Classification:  Nonparametric regression,  Brownian motion,  nonstandard likelihood asymptotics,  tests for constant mean,  62G10,  62E20,  62M10
@article{1176347753,
     author = {Barry, Daniel and Hartigan, J. A.},
     title = {An Omnibus Test for Departures from Constant Mean},
     journal = {Ann. Statist.},
     volume = {18},
     number = {1},
     year = {1990},
     pages = { 1340-1357},
     language = {en},
     url = {http://dml.mathdoc.fr/item/1176347753}
}
Barry, Daniel; Hartigan, J. A. An Omnibus Test for Departures from Constant Mean. Ann. Statist., Tome 18 (1990) no. 1, pp.  1340-1357. http://gdmltest.u-ga.fr/item/1176347753/