1School of Mathematics and statistics, Huazhong University of Science and Technology, Wuhan, Hubei, 430074, P.R. China
2Department of Mathematics, College of Science, Basra University, Basra, Iraq
Online published on 22 August, 2016.
In this paper, a class of statistical modelsis proposed, where random effects are inserted into onedimensional stochastic differential equations (SDEs) model; SDE defined N independent stochastic processes (Xi(t), t ɛ [0, Ti), i =1,…N. The drift term dependent on a random variableφi, we have been discussedtheparametric estimation of the density of the random effect φi within two kinds of mixed models. Anadditive and multiplicativerandom effect are successively considered, when φi has exponential distribution. We obtained an expression of the exact likelihood and proved the consistency and asymptotic normality of the maximum likelihood estimators.
Asymptotic normality, consistency, maximum likelihood estimator, mixed effects stochastic differential equations