TIME SERIES ANALYSIS ON STOCK MARKET FOR TEXT MINING CORRELATION OF ECONOMY NEWS

• Main Authors: Sadi Evren SEKER, Cihan MERT, Khaled Al-NAAMI, Nuri OZALP, Ugur AYAN
• Format: Article
• Language: English
• Published: Social Sciences Research Society 2014-01-01
• Series: International Journal of Social Sciences and Humanity Studies
• Online Access: http://www.sobiad.org/ejournals/journal_IJSS/arhieves/IJSS-2014_1/Sadi-Evren.pdf

This paper proposes an information retrieval methodfor the economy news. Theeffect of economy news, are researched in the wordlevel and stock market valuesare considered as the ground proof.The correlation between stock market prices and economy news is an already ad-dressed problem for most of the countries. The mostwell-known approach is ap-plying the text mining approaches to the news and some time series analysis tech-niques over stock market closing values in order toapply classification or cluster-ing algorithms over the features extracted. This study goes further and tries to askthe question what are the available time series analysis techniques for the stockmarket closing values and which one is the most suitable? In this study, the newsand their dates are collected into a database and text mining is applied over thenews, the text mining part has been kept simple with only term frequency – in-verse document frequency method. For the time series analysis part, we havestudied 10 different methods such as random walk, moving average, acceleration,Bollinger band, price rate of change, periodic average, difference, momentum orrelative strength index and their variation. In this study we have also explainedthese techniques in a comparative way and we have applied the methods overTurkish Stock Market closing values for more than a2 year period. On the otherhand, we have applied the term frequency – inversedocument frequency methodon the economy news of one of the high-circulatingnewspapers in Turkey.