JCI MODELING IN INDONESIA BASED ON INDUSTRIAL PRODUCTION INDEX WITH LOCAL POLYNOMIAL ESTIMATOR APPROACH

Abstract

The industrial sector is the leading sector that contributes the most to Indonesia's economic growth. Industry can be caused by various factors, one of which is the Jakarta Composite Index (JCI). Indonesian stock prices have a high variance that requires proper modeling. Therefore, this study uses a local polynomial nonparametric regression approach. This study aims to estimate and obtain the best JCI model based on the production index of large and medium industries using a local polynomial estimator and also knowing the accuracy of the JCI model based on the production index of large and medium industries. The data used in this study is secondary data using production index data for medium-large industries and data on the composite stock index in Indonesia in the form of Time series which were obtained through the Central Statistics Agency Publication website on the page www.bps.go.id. JCI modeling in Indonesia based on the production index of large and medium industries is most effective on local polynomials with polynomial degree two which obtains an optimal bandwidth of 7,8795 with a minimum Cross-Validation (CV) value of 163170,3 and a Mean Absolute Percentage Error (MAPE) value of 9,1%. From the MAPE value it is said that the model is good for making future predictions.