TEJOSANTOSO, CHRISTIANTO KEVIN (2023) EVALUATING COAL STOCKS WITH ANN,RANDOM FOREST AND ARIMA MODEL. Other thesis, Universitas Katolik Soegijapranata Semarang.

Preview	Text (COVER) 19.K1.0067-CHRISTIANTO KEVIN TEJOSANTOSO-COVER_a.pdf Download (764kB) | Preview
	Text (BAB I) 19.K1.0067-CHRISTIANTO KEVIN TEJOSANTOSO-BAB I_a.pdf Restricted to Registered users only Download (116kB)
	Text (BAB II) 19.K1.0067-CHRISTIANTO KEVIN TEJOSANTOSO-BAB II_a.pdf Restricted to Registered users only Download (193kB)
	Text (BAB III) 19.K1.0067-CHRISTIANTO KEVIN TEJOSANTOSO-BAB III_a.pdf Restricted to Registered users only Download (285kB)
	Text (BAB IV) 19.K1.0067-CHRISTIANTO KEVIN TEJOSANTOSO-BAB IV_a.pdf Restricted to Registered users only Download (452kB)
	Text (BAB V) 19.K1.0067-CHRISTIANTO KEVIN TEJOSANTOSO-BAB V_a.pdf Restricted to Registered users only Download (175kB)
Preview	Text (DAFTAR PUSTAKA) 19.K1.0067-CHRISTIANTO KEVIN TEJOSANTOSO-DAPUS_a.pdf Download (194kB) | Preview
	Text (LAMPIRAN) 19.K1.0067-CHRISTIANTO KEVIN TEJOSANTOSO-LAMP_a.pdf Restricted to Registered users only Download (239kB)

Abstract

Predicting the stock market has been challenging for years for so many people around the world. Previously, most predictions about stock price movement were done by applying Technical Analysis. With the rise of Machine Learning adoption, the application of machine learning has been combined with technical analysis which helped analysts and researchers to get better forecasts and results. Along with this trend, the usage of machine learning will increase as the technology keeps improving and the complexity on how to analyze the market will be even more difficult with so many variables coming up. Although there have been many papers that discuss the application of machine learning on stock prediction, there are only few papers that discuss the stock price predictions of Indonesian Market, specifically in forecasting coal stocks. In order to solve the issue,in this paper the author proposes an idea to use Artificial Neural Network (ANN) and Random Forest and ARIMA Model that will be evaluated with Mean Absolute Error (MAE),Root Mean Squarred Error (RMSE), Mean Absolute Percentage Error (MAPE) and Mean Squarred Error (MSE) as performance metrics of the prediction models. The results shows that ANN and ARIMA are suitable to forecast future coal stocks price .Random Forest has the highest error percentage among ANN and ARIMA Model. Therefore, the future research can explore further the effect of ARIMA Model and ANN in other industries

Actions (login required)

View Item