The categorization results show that a high rate of Tweets entering a cluster represents the occurrence of a new event in near future. Tweets are then categorized using the non-negative matrix factorization analysis and the distance dependent Chinese restaurant process incremental clustering. In the proposed method, the Tweets are initially preprocessed in consecutive fixed-length time windows. In this study, events are predicted through analyzing Twitter messages and examining the changes in the rate of Tweets in a specified subject. Prediction of events especially in the management of social crises can be of particular significance. We find that most added value is achieved in regimes associated with lower implied volatility, but optimal regimes vary per market sector.Įvent detection using social media analysis has attracted researchers’ attention. Lastly, we explore how our proposed approach fares throughout time by identifying four underlying market regimes in implied volatility using hidden Markov models. Further analysis shows that possible reasons for these discrepancies might be caused by either excess social media attention or low option liquidity. In doing so, we uncover that stocks in certain sectors, such as Consumer Discretionary, Technology, Real Estate, and Utilities are easier to predict than others. We study the approach on a stock universe comprised of the 165 most liquid US stocks diversified across the 11 traditional market sectors using a sizeable out-of-sample period spanning over six years. Through an ablation study, we examine the usefulness of different sources of predictors and expose the value of attention and sentiment features extracted from Twitter. In this study, we predict next-day movements of stock end-of-day implied volatility using random forests.
0 kommentar(er)
