UNS Conference Portal, The 1st International Conference on Science, Mathematics, Environment and Education 2017

Font Size: 
SPATIAL AUTOREGRESSIVE WITH A SPATIAL AUTOREGRESSIVE ERROR TERM (SAR-SAR) MODEL AND ITS PARAMETER ESTIMATION WITH TWO STAGE GENERALIZED SPATIAL LEAST SQUARE (GS2SLS) PROCEDURE
Dewi Retno Sari Saputro

Last modified: 2017-07-19

Abstract


The spatial lag regression model takes the dependence in the dependent variable of a spatial unit and the corresponding neighboring units into account, while the spatial error regression model considers spatial dependence in the error term of a spatial unit and the corresponding neighboring units. A model with two forms of dependence is expressed as autoregressive spatial model with a spatial autoregressive error term (SAR-SAR). The resulting dependence parameter estimation by Ordinary Least Squares (OLS) method provides inconsistent estimators. Therefore, we need another estimation method that produces consistent ones. The method is termed Generalized Spatial Two Stage Least Square (GS2SLS) procedure. This article seeks to prove parameter estimation of SAR-SAR model with GS2SLS.