A Dynamical System and Neural Network Perspective of Karachi Stock Exchange

Syed Nasir Danial; Syed Raheel Noor; Bilal A. Usmani; S. Jamal H. Zaidi

A Dynamical System and Neural Network Perspective of Karachi Stock Exchange

Syed Nasir Danial, Syed Raheel Noor, Bilal A. Usmani, S. Jamal H. Zaidi

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

2 Citations (Scopus)

Abstract

This study discusses the evolution of KSE-100 index returns as a dynamical system. We present application of nonlinear time-series analysis. Our results show that estimation of correlation dimension for the case of KSE-100 index returns is not possible. We further go into nonlinear analysis and construct a model of the series based on feedforward neural network with back-propagation training. We construct many neural networks and the one with Levenberg-Marquardt backpropagation is found to give slightly better results compared to ARMA/ARIMA models. Neural networks are found to be applicable in those cases when nonlinear time-series analysis is at failure.

Original language	English (UK)
Title of host publication	Communications in Computer and Information Science
Editors	D.M. Akbar Hussain, Abdul Qadeer Khan Rajput, Bhawani Shankar Chowdhry, Quintin Gee
Pages	88-99
Number of pages	12
Publication status	Published - 2009
Externally published	Yes

Publication series

Name	Communications in Computer and Information Science
Volume	20
ISSN (Print)	1865-0929

Keywords

Correlation dimension
KSE-100 index returns
Neural network
Non-linear time series analysis

Cite this

@inproceedings{cba43a6a0dac4dcfbaac00c78ed81da5,

title = "A Dynamical System and Neural Network Perspective of Karachi Stock Exchange",

abstract = "This study discusses the evolution of KSE-100 index returns as a dynamical system. We present application of nonlinear time-series analysis. Our results show that estimation of correlation dimension for the case of KSE-100 index returns is not possible. We further go into nonlinear analysis and construct a model of the series based on feedforward neural network with back-propagation training. We construct many neural networks and the one with Levenberg-Marquardt backpropagation is found to give slightly better results compared to ARMA/ARIMA models. Neural networks are found to be applicable in those cases when nonlinear time-series analysis is at failure.",

keywords = "Correlation dimension, KSE-100 index returns, Neural network, Non-linear time series analysis",

author = "Danial, \{Syed Nasir\} and Noor, \{Syed Raheel\} and Usmani, \{Bilal A.\} and Zaidi, \{S. Jamal H.\}",

year = "2009",

language = "English (UK)",

isbn = "9783540898528",

series = "Communications in Computer and Information Science",

pages = "88--99",

editor = "\{Akbar Hussain\}, D.M. and Rajput, \{Abdul Qadeer Khan\} and Chowdhry, \{Bhawani Shankar\} and Quintin Gee",

booktitle = "Communications in Computer and Information Science",

}

A Dynamical System and Neural Network Perspective of Karachi Stock Exchange. / Danial, Syed Nasir; Noor, Syed Raheel; Usmani, Bilal A. et al.
Communications in Computer and Information Science. ed. / D.M. Akbar Hussain; Abdul Qadeer Khan Rajput; Bhawani Shankar Chowdhry; Quintin Gee. 2009. p. 88-99 (Communications in Computer and Information Science; Vol. 20).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - A Dynamical System and Neural Network Perspective of Karachi Stock Exchange

AU - Danial, Syed Nasir

AU - Noor, Syed Raheel

AU - Usmani, Bilal A.

AU - Zaidi, S. Jamal H.

PY - 2009

Y1 - 2009

N2 - This study discusses the evolution of KSE-100 index returns as a dynamical system. We present application of nonlinear time-series analysis. Our results show that estimation of correlation dimension for the case of KSE-100 index returns is not possible. We further go into nonlinear analysis and construct a model of the series based on feedforward neural network with back-propagation training. We construct many neural networks and the one with Levenberg-Marquardt backpropagation is found to give slightly better results compared to ARMA/ARIMA models. Neural networks are found to be applicable in those cases when nonlinear time-series analysis is at failure.

AB - This study discusses the evolution of KSE-100 index returns as a dynamical system. We present application of nonlinear time-series analysis. Our results show that estimation of correlation dimension for the case of KSE-100 index returns is not possible. We further go into nonlinear analysis and construct a model of the series based on feedforward neural network with back-propagation training. We construct many neural networks and the one with Levenberg-Marquardt backpropagation is found to give slightly better results compared to ARMA/ARIMA models. Neural networks are found to be applicable in those cases when nonlinear time-series analysis is at failure.

KW - Correlation dimension

KW - KSE-100 index returns

KW - Neural network

KW - Non-linear time series analysis

UR - https://www.scopus.com/pages/publications/67649670431

M3 - Conference contribution

AN - SCOPUS:67649670431

SN - 9783540898528

T3 - Communications in Computer and Information Science

SP - 88

EP - 99

BT - Communications in Computer and Information Science

A2 - Akbar Hussain, D.M.

A2 - Rajput, Abdul Qadeer Khan

A2 - Chowdhry, Bhawani Shankar

A2 - Gee, Quintin

ER -

A Dynamical System and Neural Network Perspective of Karachi Stock Exchange

Abstract

Publication series

Keywords

Other files and links

Fingerprint

Cite this