LIMITED INFORMATION MAXIMUM LIKELIHOOD USING PRINCIPAL COMPONENTS AND ITS APPLICATION

    • ONISHI Haruo
    • Institute of Socio-Economic Planning, University of Tsukuba

抄録

In cases where the number of predetermined variables exceeds the sample size or strong multicollinearity occurs among excluded predetermined variables, the limited information maximum likelilhood and the least variance ratio cannot be used to estimate a simtultaneous equation system. We propose a method in which some effectively-influential principal components of excluded predetermined variables are employed as instrumental variables in place of the excluded predetermined variables. Then we demonstrate the estimation of a Cobb-Douglas agricutural production function with the proposed method.

収録刊行物

Journal of the Japanese Society of Computational Statistics   [巻号一覧]

Journal of the Japanese Society of Computational Statistics 6(2), 11-19, 1993-12  [この号の目次]

日本計算機統計学会