Non-Parametric Causality Detection: An Application to Social Media and Financial Data

TL;DR

This paper introduces a non-parametric causal inference framework to determine whether social media influences stock market returns, overcoming limitations of traditional correlation and regression analyses.

Contribution

The work presents a novel non-parametric causality detection method that does not assume specific statistical relationships, effectively controlling for multiple confounding factors.

Findings

Social media data causally influence stock returns

The method successfully distinguishes causality from correlation

Empirical results support social media's impact on market dynamics

Abstract

According to behavioral finance, stock market returns are influenced by emotional, social and psychological factors. Several recent works support this theory by providing evidence of correlation between stock market prices and collective sentiment indexes measured using social media data. However, a pure correlation analysis is not sufficient to prove that stock market returns are influenced by such emotional factors since both stock market prices and collective sentiment may be driven by a third unmeasured factor. Controlling for factors that could influence the study by applying multivariate regression models is challenging given the complexity of stock market data. False assumptions about the linearity or non-linearity of the model and inaccuracies on model specification may result in misleading conclusions. In this work, we propose a novel framework for causal inference that does not require any assumption about the statistical relationships among the variables of the study and can effectively control a large number of factors. We apply our method in order to estimate the causal impact that information posted in social media may have on stock market returns of four big companies. Our results indicate that social media data not only correlate with stock market returns but also influence them.