Early warning of banking crises using the Markov model

Applied Study on a sample of countries for the period )2000-2023)

A Dissertation Submitted to The Council of College of Administration and Economics at Karbala University, As Partial Fulfillment of the Requirements for PH.D. Degree Financial and Banking Sciences.

Submitted by

Zahraa Yousef Abbas Al- Saadi

Under The Supervision

prof. Dr. Abbas Kazim Jassim Al-Da’ami

Abstract

     The study aimed at enabling Governments and central banks to cope with banking crises by forecasting banking crises through a series of historical statements of a sample of financial and economic variables from developed and emerging countries’ economies and testing a proposed prediction model represented in the Markov model for system replacement (Markov-Switching) to reach the possibility of generating accurate predictive signals capable of predicting future crises. Forecasting models may have been and continue to be intellectual and applied debates about the relevance and preference of these models for predicting banking crises, especially after traditional warning models failed to predict the global financial crisis (2008).

So this study came to see this argument and try to solve it by testing the Markov model to switch the system in the light of the data obtained for the study sample of five countries: (United States, Canada, Brazil, Chile and Turkey) and using quarterly data for the duration (first quarter of 2000) until (Fourth quarter of 2023), using many financial, statistical and sports methods, the study concluded a number of conclusions, perhaps the most important of which is that Markov’s model of system replacement is strong as an early warning indicator for banking crisis forecasting processes. It could generate a higher probability of crisis. Furthermore, the analysis within the sample indicates that these indicators can provide an early warning signal of up to several quarters before changing the system concerned. The study has produced a number of recommendations. Perhaps most important is the need to adopt the Markov model as an early warning model as a result of its accuracy in predicting banking crises.