Early Detection Model of Currency Crisis in Malaysia Based on Real Exchange Rate Indicators Using Markov Switching

Beginning in July 1997, Malaysia had a currency balance deficit above five percent of GDP. Significant shifts in macroeconomic indices, including a fall in the real exchange rate, were closely associated with the currency collapse. This circumstance may lead to a Malaysian currency crisis. These variables must be monitored as the crisis may affect extremely adverse economic circumstances. A model for early currency crisis identification in Malaysia is developed based on the behavior of this indicator. When there is a behavioral crisis, the conditions shift from non-crisis to crisis and the indicator varies. Markov Switching integration into Generalized Autoregressive Conditional Heteroscedasticity (MS-GARCH) models provides a framework for understanding and analyzing the nuances present in a crisis. This integrated method creates a solid foundation for crisis research and provides sophisticated knowledge that surpasses traditional methods. Combining these cutting-edge statistical techniques gives us a more sophisticated grasp of the dynamics of the crisis and enables a more thorough study that surpasses the capabilities of conventional models. By using an integrated strategy, we may better foresee, manage, and lessen the impact of crisis occurrences on financial institutions and markets by utilizing a comprehensive set of instruments to untangle their intricacies. The results show that the smoothed probability value in the MS-GARCH(2,1,1) model can predict the absence of a tendency for a crisis to occur in 2023.