Causal approaches to financial markets

Vortrag

Understanding drivers for price change in financial markets clearly is of great importance for both theory and practice. Traditional quantitative methods are typically multivariate linear regression or tools like principal component analysis and offer limited insight into causal mechanisms. In particular, this type of analysis is restricted to the one history that played out. Market professionals often supplement such quantitative analysis by contextual human knowledge and their interpretation of financial, economic, and political, news, but rarely do so in a formalized way. In Special Session "Quantum Methods in Economics and Finance" organized by Emmanuel Haven Abstract of talk: Since the seminal work of John Bell causal models are used in quantum foundations to understand the constraints for hidden variable models that aim at reproducing quantum statistics. Notably, the causal approach allows to address counterfactual questions. It can be fully formalized and then used as a test to rule out competing causal models. In physics, this is possible with spatially separated entities or across time through Leggett-Garg Inequalities. In this talk, we discuss the use of such methods for practical applications in finance. In particular, we address the question whether they can give an early indication about changes in the current causal structure that is the driving force in a market. Of course, in finance causal relationships can change over time, for example through exogeneous events (e.g. shutdowns during Covid) or through changes in investor perception (e.g. changes in the view of what constitutes “sustainable” business activity and ESG compliant conduct). Accordingly, we will look at both the macro and micro level to see whether there is a possible “quantum advantage”.