Mj. Silvapulle et P. Silvapulle, A SCORE TEST AGAINST ONE-SIDED ALTERNATIVES, Journal of the American Statistical Association, 90(429), 1995, pp. 342-349
A score-type statistic, T-s, is introduced for testing H : psi = 0 aga
inst K : psi greater than or equal to 0 and more general one-sided hyp
otheses when nuisance parameters may be present; psi is a vector param
eter. The main advantages of T-s are that it requires estimation of th
e model only under the null hypothesis and that, it is asymptotically
equivalent to the likelihood ratio statistic; these are precisely the
reasons for the popularity of the score tests for testing against two-
sided alternatives. In this sense, T-s preserves the main attractive f
eatures of the classical two-sided score test. The theoretical results
are presented in a general framework where the likelihood-based score
function is replaced by an estimating function so that the test is ap
plicable even if the exact population distribution is unknown. Computa
tion of T-s is simplified by the fact that it can be computed easily o
nce the corresponding two-sided statistic has been computed. The relev
ance and simplicity of T-s are illustrated by discussing a data exampl
e in detail. This example involves an autoregressive conditional heter
oscedasticity (ARCH) model, and the objective is to test for the prese
nce of ARCH effect. The null and alternative hypotheses turn out to be
of the form H : psi = 0 and K : psi greater than or equal to 0 respec
tively where psi is the vector of ARCH parameters. In contrast to the
likelihood ratio and other equivalent forms that are currently availab
le for testing against such one-sided hypotheses, we note the followin
g: T-s is convenient to apply, because the full ARCH model need not be
estimated subject to the inequality constraint psi greater than or eq
ual to 0 and we do not need to know the exact likelihood because T-s i
s based on estimating equations rather than likelihoods.