FORECASTING SAHAM MENGGUNAKAN METODE CONVULATION NEURAL NETWORKS (CNN) – LONG SHORT-TERM MEMORY (LSTM)

PRATAMA, SHANDY and Samsuryadi, Samsuryadi and Rizqie, Muhammad Qurhanul (2022) FORECASTING SAHAM MENGGUNAKAN METODE CONVULATION NEURAL NETWORKS (CNN) – LONG SHORT-TERM MEMORY (LSTM). Undergraduate thesis, Sriwijaya University.

	Text RAMA_55201_09021181823020.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (2MB) | Request a copy
	Text RAMA_55201_09021181823020_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (4MB) | Request a copy
	Text RAMA_55201_09021181823020_0004027101_0203128701_01_front_ref.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy
	Text RAMA_55201_09021181823020_0004027101_0203128701_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (386kB) | Request a copy
	Text RAMA_55201_09021181823020_0004027101_0203128701_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (428kB) | Request a copy
	Text RAMA_55201_09021181823020_0004027101_0203128701_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (601kB) | Request a copy
	Text RAMA_55201_09021181823020_0004027101_0203128701_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (154kB) | Request a copy
	Text RAMA_55201_09021181823020_0004027101_0203128701_06.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (75kB) | Request a copy
	Text RAMA_55201_09021181823020_0004027101_0203128701_07_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (145kB) | Request a copy
	Text RAMA_55201_09021181823020_0004027101_0203128701_08_lamp.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (107kB) | Request a copy

Abstract

Stock is one of the financial instruments of the stock market that is in great demand by the public. Stock prices can change over time. There are several factors that cause changes in stock prices. Based on this, a system can predict stock price data using the Convolutional Neural Network (CNN) - Long Short-Term Memory (LSTM) method. The data used in this prediction is data for the last 2 years from Facebook and Tesla stocks. The amount of data used is 505 data, then divided into 75% training data and 25% test data. In conducting the test, each data is tested with different configuration combinations. The results of the test show that the configuration that most influences the loss results from predictions is the epoch and the amount of data used. Facebook shares with epoch 200 configuration and padding off have smaller loss values, namely RMSE: 6.72 and MAE: 5.12.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	Forecasting, CNN, LSTM, Saham
Subjects:	T Technology > T Technology (General) > T1-995 Technology (General)
Divisions:	09-Faculty of Computer Science > 55201-Informatics (S1)
Depositing User:	Shandy Pratama
Date Deposited:	17 Jan 2023 07:01
Last Modified:	17 Jan 2023 07:02
URI:	http://repository.unsri.ac.id/id/eprint/82695

Actions (login required)

View Item