Scholarly record

MULTISCALE PERSPECTIVE IN SUPPORT OF INVESTMENT DECISION-MAKING UNDER MARKET INSTABILITY

Nadya Velinova-Sokolova, Boryana Pelova

Abstract

Financial markets exhibit complex dynamic behavior, particularly during periods of instability when traditional assumptions about market efficiency tend to break down. This study builds upon an ML-based AI system developed in [1], [2], which extracts information from companies’ financial statements to model stock price directional movements. From a behavioral finance perspective, investor biases may lead to deviations from rational decision-making, especially during periods of market stress. To address this, a multiscale perspective is integrated into the AI-ML based system in order to identify patterns in financial time series across different time horizons. The aim is to also analyze the risks associated with market efficiency. The findings suggest that fundamental financial data seen through a multiscale lens can provide additional support for more informed investment decisions, without relying solely on technical or short-term signals. [1] Bogdanova, B. and Stancheva-Todorova, E., 2021, March. ML-based predictive modelling of stock market returns. In AIP Conference Proceedings (Vol. 2333, No. 1, p. 150006). AIP Publishing LLC. [2] Bogdanova, B., 2020, Applied AI in Support of Investment Decision-Making. Journal of Economic Boundaries and Transformation 1(1), pp. 49-60.

Publication details

Back to publication list