Using media-based emotion to predict commodity price

Abstract

Emotion plays a significant role in consumer decision making. We recently conducted a study to explore how media-based information of aggregated market emotion influences consumers' expected demand of commodities, and how businesses can use media-based emotion indices to predict commodities' price. We implemented time series econometrics by analyzing a fourteen year daily observations of twelve major energy and material commodities prices and five market-level emotion indices (buzz, sentiment, optimism, fear and joy). The empirical results suggest that high-arousal emotion from all groups of individuals tend to reach consensus at the market level in its effect on commodity price. The study also provides evidence that there is a short term predictive relationship between the media-based emotion indices and the following five days' commodity prices.