Stock trading and prediction using back-propagation neural network

In this paper, Back-Propagation neural network is used to make prediction on stock price. Theories of neural network, back propagation, and Levenberg-Marquardt algorithm are discussed to obtain a deeper understanding into the paper. Then, variable input information including basic information, techn...

Full description

Saved in:
Bibliographic Details
Main Author: Wang, Xiaogang
Other Authors: Wang Lipo
Format: Final Year Project
Language:English
Published: 2014
Subjects:
Online Access:http://hdl.handle.net/10356/61238
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
Description
Summary:In this paper, Back-Propagation neural network is used to make prediction on stock price. Theories of neural network, back propagation, and Levenberg-Marquardt algorithm are discussed to obtain a deeper understanding into the paper. Then, variable input information including basic information, technical indicators and index indicators are investigated to find the most robust input combinations. The impact of neural network architecture is also covered. The neural network with best performance is later tested on 8 other companies to evaluate its profit ability. In the end, the experiment obtained a promising result and proved Back-Propagation neural network’s capacity in stock prediction.