Time series modeling and designing of artifical neural network (ANN) for revenue forecasting

Artificial neural networks (ANN) have found increasing consideration in forecasting theory. However, the large numbers of parameters that must be selected to develop ANN forecasting model have meant that the design process still involves much trial and error. The objective of this study is to invest...

Full description

Saved in:

Bibliographic Details
Main Author:	Mohd. Yusof, Norfadzlia
Format:	Thesis
Language:	English
Published:	2005
Subjects:	QA75 Electronic computers Computer science
Online Access:	http://eprints.utm.my/id/eprint/4799/1/NorfadzliaMohdYusofMFSKSM2006.pdf
Tags:	Add Tag No Tags, Be the first to tag this record!

id	my-utm-ep.4799
record_format	uketd_dc
spelling	my-utm-ep.47992018-02-28T06:47:44Z Time series modeling and designing of artifical neural network (ANN) for revenue forecasting 2005-06 Mohd. Yusof, Norfadzlia QA75 Electronic computers. Computer science Artificial neural networks (ANN) have found increasing consideration in forecasting theory. However, the large numbers of parameters that must be selected to develop ANN forecasting model have meant that the design process still involves much trial and error. The objective of this study is to investigate the effect of applying different number of input nodes, activation functions and pre-processing techniques on the performance of backpropagation (BP) network in time series revenue forecasting. In this study, several pre-processing techniques are presented to remove the non-stationary in the time series and their effect on ANN model learning and forecast performance are analyzed. Trial and error approach is used to find the sufficient number of input nodes as well as their corresponding number of hidden nodes which obtain using Kolmogorov theorem. This study compares the used of logarithmic function and new proposed ANN model which combines sigmoid function in hidden layer and logarithmic function in output layer, with the standard sigmoid function as the activation function in the nodes. A cross-validation experiment is employed to improve the generalization ability of ANN model. From the empirical findings, it shows that an ANN model which consists of small number of input nodes and smaller corresponding network structure produces accurate forecast result although it suffers from slow convergence. Sigmoid activation function decreases the complexity of ANN and generates fastest convergence and good forecast ability in most cases in this study. This study also shows that the forecasting performance of ANN model can considerably improve by selecting an appropriate pre-processing technique. 2005-06 Thesis http://eprints.utm.my/id/eprint/4799/ http://eprints.utm.my/id/eprint/4799/1/NorfadzliaMohdYusofMFSKSM2006.pdf application/pdf en public masters Universiti Teknologi Malaysia, Faculty of Computer Science and Information System Faculty of Computer Science and Information System
institution	Universiti Teknologi Malaysia
collection	UTM Institutional Repository
language	English
topic	QA75 Electronic computers Computer science
spellingShingle	QA75 Electronic computers Computer science Mohd. Yusof, Norfadzlia Time series modeling and designing of artifical neural network (ANN) for revenue forecasting
description	Artificial neural networks (ANN) have found increasing consideration in forecasting theory. However, the large numbers of parameters that must be selected to develop ANN forecasting model have meant that the design process still involves much trial and error. The objective of this study is to investigate the effect of applying different number of input nodes, activation functions and pre-processing techniques on the performance of backpropagation (BP) network in time series revenue forecasting. In this study, several pre-processing techniques are presented to remove the non-stationary in the time series and their effect on ANN model learning and forecast performance are analyzed. Trial and error approach is used to find the sufficient number of input nodes as well as their corresponding number of hidden nodes which obtain using Kolmogorov theorem. This study compares the used of logarithmic function and new proposed ANN model which combines sigmoid function in hidden layer and logarithmic function in output layer, with the standard sigmoid function as the activation function in the nodes. A cross-validation experiment is employed to improve the generalization ability of ANN model. From the empirical findings, it shows that an ANN model which consists of small number of input nodes and smaller corresponding network structure produces accurate forecast result although it suffers from slow convergence. Sigmoid activation function decreases the complexity of ANN and generates fastest convergence and good forecast ability in most cases in this study. This study also shows that the forecasting performance of ANN model can considerably improve by selecting an appropriate pre-processing technique.
format	Thesis
qualification_level	Master's degree
author	Mohd. Yusof, Norfadzlia
author_facet	Mohd. Yusof, Norfadzlia
author_sort	Mohd. Yusof, Norfadzlia
title	Time series modeling and designing of artifical neural network (ANN) for revenue forecasting
title_short	Time series modeling and designing of artifical neural network (ANN) for revenue forecasting
title_full	Time series modeling and designing of artifical neural network (ANN) for revenue forecasting
title_fullStr	Time series modeling and designing of artifical neural network (ANN) for revenue forecasting
title_full_unstemmed	Time series modeling and designing of artifical neural network (ANN) for revenue forecasting
title_sort	time series modeling and designing of artifical neural network (ann) for revenue forecasting
granting_institution	Universiti Teknologi Malaysia, Faculty of Computer Science and Information System
granting_department	Faculty of Computer Science and Information System
publishDate	2005
url	http://eprints.utm.my/id/eprint/4799/1/NorfadzliaMohdYusofMFSKSM2006.pdf
_version_	1747814547817758720

Time series modeling and designing of artifical neural network (ANN) for revenue forecasting

Similar Items