TRAP@NCI

Analyzing Historical Stock Market Data to Determine if a Correlation Exists Between Major Stock Market Indexes and if Time Series is Sufficient to Make Predictions

Lafayette-Madden, Anicia (2016) Analyzing Historical Stock Market Data to Determine if a Correlation Exists Between Major Stock Market Indexes and if Time Series is Sufficient to Make Predictions. Masters thesis, Dublin, National College of Ireland.

[img]
Preview
PDF (Master of Science)
Download (1MB) | Preview
[img]
Preview
PDF (Configuration File)
Download (1MB) | Preview

Abstract

Time series for forecasting stock market prices has become extremely popular over the last few decades. Both stand-alone time series algorithms and hybrids have been successfully implemented, with several papers documenting their positive results. Correlation and volatility across the major markets have also seen quite a number of works conducted with positive results being documented. This research has examined the existence of a correlation between major stock market prices and the effectiveness of time series for forecasting future prices. A cross-correlation matrix and ARIMA time series algorithm for forecasting were implemented. The results, while proving correlation exists between the major stock markets, was inconclusive when examining the percentage change in US GDP growth and its possible effect on the change in correlation. Time series results were unexpected, however, this researcher strongly believes it is indeed effective in making stock market price predictions.

Item Type: Thesis (Masters)
Subjects: Q Science > QA Mathematics > Electronic computers. Computer science
T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science

H Social Sciences > HG Finance > Investment > Stock Exchange
Divisions: School of Computing > Master of Science in Data Analytics
Depositing User: CAOIMHE NI MHAICIN
Date Deposited: 03 Dec 2016 11:17
Last Modified: 03 Dec 2016 14:55
URI: http://trap.ncirl.ie/id/eprint/2487

Actions (login required)

View Item View Item