Sentiment in the Forex Markets

Important results from the project

The academic goal is to increase understanding of the function and dysfunction of financial markets. More specifically, we intend to investigate how market sentiment affects the behavior and decision-making of actors in the currency markets, and how this in turn affects the functioning of the markets themselves. The business goal is to use our measures of market sentiment to predict price changes in the market. The purpose and goals have been achieved.

Expected long term effects

The expected academic results are a number of research articles that increase our understanding of how sentiment affects actors in financial markets and how this affects the functioning of the markets themselves. The outcome is a doctoral thesis and several working papers in the field. Results for the business sector are i) a trading algorithm based on sentiment information from text sources and the order book, ii) an increased understanding of how sentiment can continuously be used to construct future trading algorithms, and iii) optimized methods for finding market anomalies.

Approach and implementation

The main applicant has overall responsibility for the project. The project is divided into two tracks, the first of which deals with measuring sentiment from the order book is led by Yana Petrova. The second track, which measures sentiment from text, is led by Hossein Asgharian. Throughout the project, CA will collect and process data, and CA will also participate in the research and implement and analyze results through a "live" trading system. These tests will provide feedback to further develop the research. The project was carried out according to plan.