Exact initial conditions for maximum likelihood estimation of state space models with stochastic inputs

被引：6

作者：

Casals, J ^{[1
]}

Sotoca, S ^{[1
]}

机构：

[1] Univ Complutense, Dept Fundamentos Analisis Econ Econ Cuantitativa, Fac C Econ & Empresariales, Madrid 28223, Spain

来源：

ECONOMICS LETTERS | 1997年 / 57卷 / 03期

关键词：

initial conditions; stationarity; Kalman filter; state space models;

D O I：

10.1016/S0165-1765(97)00180-8

中图分类号：

F [经济];

学科分类号：

02 ;

摘要：

We derive exact expressions for the conditional mean and variance of the initial state of a State Space system with stochastic inputs, under stationarity or non-stationarity. These results provide an initialization method to obtain maximum likelihood estimates of the parameters. (C) 1997 Elsevier Science S.A.

引用

页码：261 / 267

页数：7

共 18 条

[1] COVARIANCE-MATRIX COMPUTATION OF THE STATE VARIABLE OF A STATIONARY GAUSSIAN PROCESS [J].