Modelling Cambodia’s Foreign Exchange Rate Dynamics: A Markov-Switching Autoregressive Model

Abstract

The analysis of the daily fluctuations in the foreign exchange rate between the Khmer Riel and the US dollar was performed utilizing a Markov- switching autoregressive model. This study covered a time frame from January 04, 2005, to August 22, 2024, encompassing a total of 4989 days of data. The research findings revealed that the MS(2)-AR(1) model emerged as the most appropriate model for the analysis. The empirical results from both state 1 and state 2 models demonstrated that the intercept term, along with the AR(1) component, exerted a statistically significant positive effect on the exchange rate at a 1% significance level. Furthermore, the intercept term, which represents the average exchange rate, along with the volatility of the exchange rate in the state 2 model, was found to be higher than that observed in the state 1 model. The analysis of the probability transition matrix indicated that there was a 15.53% likelihood for FX to transition from state 2 to state 1. In contrast, the probability of FX departing from state 1 and returning to state 2 was recorded at 29.34%. Additionally, the chance of FX maintaining its position in state 1 was assessed to be 70.66%, while the probability of it remaining in state 2 was significantly higher at 84.47%.