The Design of an Agent-Based System for Capital Markets

Marius PETRESCU¹*, Mădălina CUC², Ionica ONCIOIU³, Anca-Gabriela PETRESCU⁴,
Florentina-Raluca BÎLCAN⁴, Mihai PETRESCU⁴
¹ Romanian Academy of Scientists, 54 Independence Splendor, 030167, Bucharest, Romania
profdrmpetrescu@yahoo.com (*Corresponding author)
² Mihai Viteazul National Intelligence Academy, Odaii 20-22, Bucharest, 075100, Romania
madalina.cuc@animv.ro
³ Titu Maiorescu University, 189 Calea Vacaresti Street, Bucharest, 040051, Romania
ionica.oncioiu@prof.utm.ro
⁴ Valahia University of Targoviste, 2 Carol I Blvd, Targoviste, 130024, Romania
anca.petrescu@valahia.ro; raluca.bilcan@valahia.ro; mihai_tina@yahoo.com

Abstract: The trading risk management implies analysing several types of risks: capital, market, liquidity, insolvency, business, credit, operational or financial risk. Trading models and techniques must be seen as tools that can provide an informed manager with useful insights, so they are indispensable in an increasingly integrated and sophisticated market. The main aim of this study is the conceptualization of a generic intelligent agent, based on a neural network and an agent system, applicable to a trading system of the listed companies. The results show that this model can contribute to reducing the transactional risk on the capital market and can offer solutions to improve the managerial decisions.

Keywords: Artificial neural networks, Multi‐agent systems, Transaction databases, Capital markets, Validation.

>FULL TEXT: PDF

Marius PETRESCU, Mădălina CUC, Ionica ONCIOIU, Anca-Gabriela PETRESCU, Florentina-Raluca BÎLCAN, Mihai PETRESCU, The Design of an Agent-Based System for Capital Markets, Studies in Informatics and Control, ISSN 1220-1766, vol. 29(3), pp. 293-306, 2020. https://doi.org/10.24846/v29i3y202003