Please use this identifier to cite or link to this item: https://hdl.handle.net/10356/140538
Title: Assess edRVFL in stock market price forecasting
Authors: Ding, Yeqiao
Keywords: Engineering::Electrical and electronic engineering
Issue Date: 2020
Publisher: Nanyang Technological University
Abstract: In a fast-growing society with economic boosting up, it is commonly seen that people without any financial background get themselves actively involved in the stock market. The stock market is known for its dramatic changes over an extremely short period of time. It is as challenging as it takes to foresee the tendency of the stock price. There are several established approaches for predicting stock price trends such as, Random Vector Functional Link ( RVFL), ,ensemble deep network(edRVFL), Artificial neural networks (ANN) and Convolutional Neural Network(CNN) . These approaches are meant to predict the stock price variation as accurate as possible, but the accuracy rate is not yet satisfactory, as the mass stock data are in high complexity, classifiers often fail to help investor maximize their profits. This study attempts to evaluate the accuracy rate of edRVFL through experiments based on 10 stocks datasets within the last 10 years. Among the 10 chosen stock, all of them will be tested and discussed in full details through all 6 different classifiers of edRVFL. An insight with test results of how window size and layer and horizon affecting the forecast results is given in this paper. Further experiments were conducted to explore the degree of improvement when adding preprocess like EMPIRICAL MODE DECOMPOSITION DISCRETE WAVELET TRANSFORM and Market indicator on edRVFL.
URI: https://hdl.handle.net/10356/140538
Fulltext Permission: restricted
Fulltext Availability: With Fulltext
Appears in Collections:EEE Student Reports (FYP/IA/PA/PI)

Files in This Item:
File Description SizeFormat 
Assess edRVFL in Stock Market Price Forecasting.pdf
  Restricted Access
1.65 MBAdobe PDFView/Open

Page view(s)

218
Updated on May 15, 2022

Download(s) 50

17
Updated on May 15, 2022

Google ScholarTM

Check

Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.