Extracting predictive information from heterogeneous data streams using Gaussian Processes

Abstract

Abstract. Financial markets are notoriously complex environments, presenting vast amounts of noisy, yet potentially informative data. We consider the problem of forecasting financial time series from a wide range of information sources using online Gaussian Processes with Automatic Relevance Determination (ARD) kernels. We measure the performance gain, quantified in terms of Normalised Root Mean Square Error (NRMSE), Median Absolute Deviation (MAD) and Pearson correlation, from fusing each of four separate data domains: time series technicals, sentiment analysis, options market data and broker recommendations. We show evidence that ARD kernels produce meaningful feature rankings that help retain salient inputs and reduce input dimensionality, providing a framework for sifting through financial complexity. We measure the performance gain from fusing each domain's heterogeneous data streams into a single probabilistic model. In particular our findings highlight the critical value of options data in mapping out the curvature of price space and inspire an intuitive, novel direction for research in financial prediction.

Social media data reveals signal for public consumer perceptions

“…where r = |x -x'|. This covariance function is a product of an exponential and a polynomial of order 1, and is suited for learning non-smooth behavior, such as those exhibited by financial timeseries [9].…”

StockTwits Classified Sentiment and Stock Returns

“…However, it is more difficult to disentangle noise from relevant tweets in Twitter than in other more focused social media. Results from Ghoshal and Roberts (2016) show that StockTwits is significantly more informative than Twitter data. This is not surprising as StockTwits is a finance-focused platform whereas Twitter also captures irrelevant opinions on a wide range of non-finance related matters.…”