SPURIOUS REGRESSION ANALYSIS ON TIME SERIES DATA FROM FACTORS AFFECTING INDONESIAN HUMAN DEVELOPMENT INDEXS IN 1990 – 2017

Abstract

In a spurious regression conditions occur linear regression equations that are not stationary on the mean and variance. If the variables are not stationary, there will be cointegration, so it can be concluded that there is a long-term equilibrium relationship between the two research variables and in the short term there is a possibility of an imbalance, so to overcome it in this study using the Error Correction Model. The purpose of this study is to apply a cointegration test to see whether there is a long-term non-equilibrium relationship between the time series between the Human Development Index and life expectancy at birth, average school year for adults aged 25 years and over and gross national income per capita. The data used in this study are time series data between 1990-2017. The statistical management is carried out using Eviews 10. Based on the results obtained, it was concluded that 81.7% and it can be said that the types of independent variables included in the model are already good, because only 18.3% of the diversity of the dependent variable is influenced by the independent variables outside this research model. Keywords: spurious regression, stationary, cointegration, error correction model, equilibrium