Br. Clarke et al., LOCAL ASYMPTOTIC THEORY FOR MULTIPLE SOLUTIONS OF LIKELIHOOD EQUATIONS, WITH APPLICATION TO A SINGLE-ION CHANNEL MODEL, Scandinavian journal of statistics, 20(2), 1993, pp. 133-146
This paper investigates local asymptotic theory when there are multipl
e solutions of the likelihood equations due to non-identifiability ari
sing as a consequence of incomplete information. Local asymptotic resu
lts extend in a stochastic way the notion of a Frechet expansion, whic
h is then used to prove asymptotic normality of the maximum likelihood
estimator under certain conditions. The local theory, which provides
simultaneous consistency and asymptotic normality results, is applied
to a bivariate exponential model exhibiting non-identifiability. This
statistical model arises as an approximation to the distribution of ob
servable sojourn-times in the states of a two-state Markov model of a
single ion channel and the non-identifiability is a consequence of an
inability to record sojourns less than some small detection limit. For
each of two possible estimates, confidence intervals are constructed
using the asymptotic theory. Although it is not possible to decide bet
ween the two estimates on the basis of a single realization, an approp
riate solution of the likelihood equations may be found using the addi
tional information contained in data based on two different detection
limits; this is investigated numerically and by simulation.