Implementasi Algoritma Regresi Linear untuk Memprediksi Harga Emas

Implementation of the Linear Regression Algorithm for Predicting Gold Prices

Abstract

This study aims to predict gold prices using several independent variables, including the silver exchange rate (SLV), the S&P 500 index (SPX), the United States Oil Fund (USO) stock exchange rate, and the Euro (EUR) to United States dollar (USD) exchange rate. The data used in this study is secondary data sourced from "Gold Price Data," comprising a total of 2290 observations and 7 columns. The method employed is regression, which is a technique for building predictive models based on given input values. The prediction results are evaluated based on the Root Mean Square Error (RMSE) value, where a smaller RMSE indicates better accuracy. The study's results show that the single-variable model has an accuracy of 73%, while the multi-variable model has an accuracy of 84%. To improve prediction accuracy, this study recommends using alternative predictive models and improving the dataset division to ensure a more representative distribution. This research not only contributes to gold price prediction but also to the development of more accurate predictive models by utilizing relevant economic variables.

Keywords: gold price prediction, regression, silver exchange rate, S&P 500 index, RMSE.