INVESTIGATING PREDICTIVE POWER OF STOCK MICRO BLOG SENTIMENT IN FORECASTING FUTURE STOCK PRICE DIRECTIONAL MOVEMENT

摘要: This study attempts to discover and evaluate the predictive power of stock micro blog sentiment on future price directional movements. We construct a set robust models based analysis data mining algorithms. Using 72,221 postings for 1909 tickers 3874 distinct authors, our reveals not only that sentiments do have simple marketadjusted returns respectively, but also this accuracy is consistent with underreaction hypothesis observed in behavioral finance. establish its succinctness, high volume real-time features over Furthermore, provides support model irrational investor sentiment, recommends complimentary investing approach using user-generated content validates an instrument may contribute monetization schemes Virtual Investing Communities.