Open AccessRegression Models for Spatial Data: An Example from Gross Domestic Regional Bruto in Province Central Java
Author(s) -
Abdul Karim,
Akhmad Faturohman,
Suhartono Suhartono,
Dedy Dwi Prastyo,
Budi Manfaat
Publication year - 2017
Publication title -
jurnal ekonomi pembangunan/jurnal ekonomi pembangunan
Language(s) - English
Resource type - Journals
eISSN - 2460-9331
pISSN - 1411-6081
DOI - 10.23917/jep.v18i2.4660
Subject(s) - spillover effect , java , lag , spatial analysis , econometrics , spatial econometrics , distributed lag , spatial dependence , regression analysis , computer science , economics , statistics , mathematics , macroeconomics , computer network , programming language
The important role of a region's transportation infrastructure strongly affects the economic growth of the region and tends to affect the surrounding areas. The effect is called spillover effect. The aim of the research was to recognize the direct effect and spillover effect (indirect) of transportation infrastructure on the economic growth in Central Java. To identify the spillover effects, it is necessary to recognize the different characteristics of each region which have the implications on the various transportation infrastructures at each region in Central Java. Therefore, the spatial modeling was conducted. In this study, the spatial modeling employed was Spatial Durbin Error Model (SDEM). The SDEM is another form of Spatial Error Model (SEM). It does not allow for lag effects of endogenous variables, but it allows for spatial error and spatial lag on exogenous variables in which it simplifies the interpretations on direct effects and spillover effect. According to SDEM estimates, the transportation infrastructures at the districts/municipalities in Central Java had no significant effect on the outputs at each region where the infrastructures were located and their neighboring districts/cities