Quasi-maximum likelihood estimation method on spatial autoregressive fixed effect for modeling the economic growth of ASEAN countries

Abstract

Economic growth is an important thing in a country's economic activities because it can be one indicator of the success of a country's economic activities. A country's economic growth is calculated based on the Gross Domestic Product (GDP) at constant prices. Some variables that are thought to influence GDP include energy consumption, trade openness, financial development, and urban population. The ASEAN Economic Community involves trade activities that are portrayed by exports and imports, thus signaling a link between ASEAN countries. The model that can be used to analyze the relationships that occur between countries is the spatial econometrics approach. The data that used in the spatial model in the form of cross-section data. On the other hand, the number of ASEAN countries is only ten, so the panel data is used in this study. Data was collected for the period 2008-2014, while the method used in this study was Quasi-Maximum Likelihood Estimation applied to spatial panel data. The results obtained show that the best model formed with QMLE test statistics with sandwich-HC covariance has the smallest CIC value of 0.0581 obtained by three independent variables that influence GDP per capita, namely energy consumption, financial development, and urban population. Based on three significant variables it is proven that there is a correlation between locations indicated by significant spatial autoregressive coefficients in the fixed-effects panel model of 0,5177. Then the value of the elasticity coefficient for the variable energy consumption, financial development, and urban population are 0,1029; 0.1433; and 0,2693.