This research aims to determine the indicators of currency crises in Vietnam. By using

the monthly data, this research attempts to establish which economic variables are more

useful in predicting currency crises and to increase the predictability of these crises. We

use time-series data from January 2001 to September 2022 to analyze and estimate the

model probability of warning of a potential currency crisis in Vietnam with an accuracy

rate of 88.89%, based on previous research papers and the use of exchange market

pressure index (EMP) and logit model. The results of the research show that the

indicators that have the strongest predictive power across the entire model are export,

real interest rate, inflation, real exchange rate, interest rate spread, net foreign assets,

domestic credit, and the money supply M2 in this order.