Predicting Stock Trading Volume through Social Media Data
Files
Authors
Chen, Yeqing
Issue Date
2016-04-01
Type
Presentation
Language
en_US
Keywords
Data analysis , Data mining , Social media , Stock trading
Alternative Title
Abstract
Social media plays a big role and can profoundly affect individual behavior and decision-making. This project is to predict the stock trading volume trend from multiple online sources. I chose Facebook to demonstrate the relationship between the stock trading volume and social media data. I used Twitter API to obtain the number of tweets, the number of “retweet” and “favorite” of twitter users. Support vector machine Model(SVM) is used in this project. The daily stock volume data is from Yahoo! Finance, Standard & Poor’s 500 indexes (S&P 500) during the period of July 2015 to December 2015. Through social media data, I analyze the tweet amount and the response form public. This way, I generated the percentage of impact rate on trading volume through social media data. It would be helpful to investment companies to predict the trend of stock market.