Naslov
Prediction of Financial Crisis in Croatia – Traditional and New Approaches

Autori
Geček Tuđen, Ivana ; Brzić, Barbara

Vrsta, podvrsta i kategorija rada
Sažeci sa skupova, sažetak, znanstveni

Skup
The 7th international scientific conference ERAZ 2021

Mjesto i datum
Beograd, Srbija, 27.05.2021

Vrsta sudjelovanja
Ostalo

Vrsta recenzije
Recenziran

Ključne riječi
Financial crises, Logit, Signaling approach, Machine learning, Early warning system

Sažetak
Financial crises are events in the economy that still pose a challenge for correct and precise forecasting. They require timely determination of accurate indicators of the crisis, choosing the right method for processing the information the indicators provide, and producing extremely reliable estimates and predictions because - on the one hand, false alarming is costly and not politically popular for policymakers, and yet if the crisis is not recognized on time, the economy takes years to recover from the crisis and in today’s globalized world, the effect of the crisis is usually widely spilled outside of the borders of the country from which it originated. While for developed countries there are many papers and discussions on the subject, for developing countries such as Croatia and comparable countries, both data and methodology is in its infancy. This article provides the comparison of the traditional prediction models (such as probit and logit and signaling model), and the new methods which are improved by the use of new technologies and techniques (such as machine learning), with a focus on Croatia in the setting of comparable countries in the EU. Based on detailed research, this article proposes methods that could most reliably and promptly predict the financial crisis in Croatia. The results of this research which is discussed in the paper, are particularly important and applicable in the period following the current pandemic crisis, especially taking into account the fact that developing countries such as Croatia are usually most affected by financial crises and have longer recovery times.

Izvorni jezik
Engleski

Znanstvena područja
Matematika, Ekonomija

POVEZANOST RADA