Prediksi Harga Saham Indonesia pada Masa Covid-19 Menggunakan Regresi Pohon Keputusan

Abstract

Stock price prediction is an interesting area of data mining. Many variables affect stock prices. Stock prices fluctuate, especially in the Covid-19 era which has an impact on the economy. This study aims to predict the stock price of Indonesian telecommunications, especially in the covid19 era. The object of this research is the financial statements of companies listed on the Indonesia Stock Exchange, where data related to stock prices of telecommunications companies are used as samples in the data processing of this study. In this study, the regression techniques that will be used are Multiple Linear Regression, Support Vector Regression, Decision Tree Regression, and K-Nearest Regression. The results showed that the fundamental data and stock prices have a relationship. High correlation coefficient resulted from Decision Tree Regression and K-Nearest Regression. Decision Tree Regression produces good results on the Train Test Split and KFold Cross Validation data, 2.99% and 2.98% repeatedly. It can conclude that despite the fact that the pandemic scenario has had a significant impact on the stock market, the study's findings suggest that fundamental data and stock prices have a relationship. Decision Tree Regression and K-Nearest Regression both produced high correlation coefficients.