2018/05/15 by Huiwen Wang, Shan Lu, Wang, Huiwen +3
Computer Science · Decision Sciences · Economics, Econometrics and Finance · #Applications (stat.AP) #Complex Systems and Time Series Analysis #Computational Engineering #Computational Finance (q-fin.CP) #FOS: Computer and information sciences #FOS: Economics and business #Finance #Stock Market Forecasting Methods #Time Series Analysis and Forecasting #and Science (cs.CE)
paper · pdf · doi:10.48550/arxiv.1805.05617
openalex publication_date 2018/05/15 · openalex created_date 2021/02/01 · openalex updated_date 2026/07/28
The increasing richness in volume, and especially types of data in the\nfinancial domain provides unprecedented opportunities to understand the stock\nmarket more comprehensively and makes the price prediction more accurate than\nbefore. However, they also bring challenges to classic statistic approaches\nsince those models might be constrained to a certain type of data. Aiming at\naggregating differently sourced information and offering type-free capability\nto existing models, a framework for predicting stock market of scenarios with\nmixed data, including scalar data, compositional data (pie-like) and functional\ndata (curve-like), is established. The presented framework is\nmodel-independent, as it serves like an interface to multiple types of data and\ncan be combined with various prediction models. And it is proved to be\neffective through numerical simulations. Regarding to price prediction, we\nincorporate the trading volume (scalar data), intraday return series\n(functional data), and investors' emotions from social media (compositional\ndata) through the framework to competently forecast whether the market goes up\nor down at opening in the next day. The strong explanatory power of the\nframework is further demonstrated. Specifically, it is found that the intraday\nreturns impact the following opening prices differently between bearish market\nand bullish market. And it is not at the beginning of the bearish market but\nthe subsequent period in which the investors' "fear" comes to be indicative.\nThe framework would help extend existing prediction models easily to scenarios\nwith multiple types of data and shed light on a more systemic understanding of\nthe stock market.\n