Twitter Analysis - Do Twitter Sentiments Correlate to Changes of Swedish Stock Prices?

Abstract: Stock market prediction is a problem that has undergone extensive research with many approaches and methods, such as mathematical models, machine learning methods et cetera. Another interesting approach is sentiment analysis, an approach takes the public opinion into account when predicting stock market prices. This method, combined with some machine learning techniques have proven effective when it comes to predict stock prices. This study determines whether this method is usable on demographics where information on public opinion does not come in abundance, in this instance the demographic of people who speak Swedish. The public sentiment is gathered by analyzing public opinion found on Twitter, and hourly stock prices for three companies were gathered. This data was combined and linear regression was performed to see if there does indeed exist a possible correlation between these data sets. The results showed that there does not appear to be a linear relation between public sentiment and changes in stock prices. The mean squared error of the data points indicate that the data points deviate to much from the regression line for the regression line to be usable as an accurate model. The limited amount of data on public sentiment led to the conclusion that Swedish Twitter flow is not usable as a source for extracting reliable information on public sentiment to be analyzed by any model.