A time series analysis of Australian stock data

Research output: Contribution to journalArticleResearchpeer-review

Abstract

This paper provides an empirical time series analysis of the daily returns of six Australian Indices, them being, All Ordinaries (Allords); Transportation, Mining, Banking, 20 Leaders and Industrials over the period 3 October 1994 to 30 September 1996. We have analysed the date using linear regression model and Box-Jenkins auto regressive integrated moving average (ARIMA) model. We have also investigated possibility of seasonality over five days of the week (Monday to Friday when the transactions takes place). Whereas no seasonal pattern is observed, Box-Jenkins approach is also not found suitable for forecasting these indices. However, in some cases we observed that auto correlation function (ACF) and partial auto correlation function (PACF) are significant at lag one which contradicts the random walk hypothesis. Overall regression model is found to give best forecast.

Original languageEnglish
Pages (from-to)103-116
Number of pages14
JournalJournal of Statistics and Management Systems
Volume2
Issue number2-3
DOIs
Publication statusPublished - 14 Jun 1999

Fingerprint

Autocorrelation
Time series analysis
Box-Jenkins
Integrated
Linear regression model
Lag
Seasonality
Random walk hypothesis
Regression model
Moving average
Banking

Cite this

@article{bb96ef0aba49494bb705c31c2abf2916,
title = "A time series analysis of Australian stock data",
abstract = "This paper provides an empirical time series analysis of the daily returns of six Australian Indices, them being, All Ordinaries (Allords); Transportation, Mining, Banking, 20 Leaders and Industrials over the period 3 October 1994 to 30 September 1996. We have analysed the date using linear regression model and Box-Jenkins auto regressive integrated moving average (ARIMA) model. We have also investigated possibility of seasonality over five days of the week (Monday to Friday when the transactions takes place). Whereas no seasonal pattern is observed, Box-Jenkins approach is also not found suitable for forecasting these indices. However, in some cases we observed that auto correlation function (ACF) and partial auto correlation function (PACF) are significant at lag one which contradicts the random walk hypothesis. Overall regression model is found to give best forecast.",
author = "Kuldeep Kumar",
year = "1999",
month = "6",
day = "14",
doi = "10.1080/09720510.1999.10700993",
language = "English",
volume = "2",
pages = "103--116",
journal = "Journal of Statistics and Management Systems",
issn = "0972-0510",
publisher = "IOS Press",
number = "2-3",

}

A time series analysis of Australian stock data. / Kumar, Kuldeep.

In: Journal of Statistics and Management Systems, Vol. 2, No. 2-3, 14.06.1999, p. 103-116.

Research output: Contribution to journalArticleResearchpeer-review

TY - JOUR

T1 - A time series analysis of Australian stock data

AU - Kumar, Kuldeep

PY - 1999/6/14

Y1 - 1999/6/14

N2 - This paper provides an empirical time series analysis of the daily returns of six Australian Indices, them being, All Ordinaries (Allords); Transportation, Mining, Banking, 20 Leaders and Industrials over the period 3 October 1994 to 30 September 1996. We have analysed the date using linear regression model and Box-Jenkins auto regressive integrated moving average (ARIMA) model. We have also investigated possibility of seasonality over five days of the week (Monday to Friday when the transactions takes place). Whereas no seasonal pattern is observed, Box-Jenkins approach is also not found suitable for forecasting these indices. However, in some cases we observed that auto correlation function (ACF) and partial auto correlation function (PACF) are significant at lag one which contradicts the random walk hypothesis. Overall regression model is found to give best forecast.

AB - This paper provides an empirical time series analysis of the daily returns of six Australian Indices, them being, All Ordinaries (Allords); Transportation, Mining, Banking, 20 Leaders and Industrials over the period 3 October 1994 to 30 September 1996. We have analysed the date using linear regression model and Box-Jenkins auto regressive integrated moving average (ARIMA) model. We have also investigated possibility of seasonality over five days of the week (Monday to Friday when the transactions takes place). Whereas no seasonal pattern is observed, Box-Jenkins approach is also not found suitable for forecasting these indices. However, in some cases we observed that auto correlation function (ACF) and partial auto correlation function (PACF) are significant at lag one which contradicts the random walk hypothesis. Overall regression model is found to give best forecast.

U2 - 10.1080/09720510.1999.10700993

DO - 10.1080/09720510.1999.10700993

M3 - Article

VL - 2

SP - 103

EP - 116

JO - Journal of Statistics and Management Systems

JF - Journal of Statistics and Management Systems

SN - 0972-0510

IS - 2-3

ER -