(in Theory and Methods)
The Power of the Likelihood Ratio Test of Location in Nonlinear
Regression Models
A. R. Gallant
Journal of the American Statistical Association, Vol. 70,
No. 349. (Mar., 1975), pp. 198-203.
Abstract
The Likelihood Ratio Test statistic, T, is considered for
the hypothesis
H: q = q0
against
A: q ¹ q0
in the nonlinear regression model
y = f(x, q) + e
with normal errors and unknown variance.
The distribution function of a random
variable X such that n·(T - X)
converges in probability to zero is derived.
Using X to approximate T, the power of
the Likelihood Ratio Test is tabulated for
selected sample sizes and departures from
the null hypothesis.
The adequacy of the approximation
of T by X is investigated in a
Monte-Carlo study.