Answers are grounded in this paper and cite their sources.
The existing models for predicting the Ruble exchange rate against the Chinese Yuan (CNY), Euro (EUR), British Pound (GBP), and United States Dollar (USD) may prove inadequate considering the Russia-Ukraine war. This study employs the Autoregressive Integrated Moving Average (ARIMA), the Evidential Neural Network with Gaussian Random Fuzzy Numbers (EVNN), and the Artificial Neural Network optimised with Particle Swarm Optimisation (ANN-PSO), as well as hybrids of ARIMA and EVNN (ARIMA-EVNN) and ARIMA and ANN-PSO (ARIMA-ANN-PSO), to predict CNY, EUR, GBP, and USD. When compared with ARIMA, for the training sample, ARIMA-ANN-PSO enhances the accuracy of ARIMA by 66.89%, 66.20%, 72.97%, and 66.89% for CNY, EUR, GBP, and USD respectively, while ARIMA-EVNN improves the accuracy of ARIMA for CNY by 66.21%, EUR by 78.87%, GBP by 82.43%, and USD by 80.33%. For the testing sample, the ARIMA-ANN-PSO model enhances the predictive accuracy of ARIMA by 65.60%, 64.39%, 45.74%, and 55.28% respectively, while the ARIMA-EVNN model improves the accuracy by 58.73% for CNY, 80.30% for EUR, 83.72% for GBP, and 86.18% for USD. The Russia-Ukraine war dummy was included to capture structural changes in Ruble exchange rate dynamics. Including war-related information does not change the accuracy of ARIMA model for CNY likely because China has not imposed sanctions on Russia, but improves it for EUR, GBP, and USD. However, the accuracy of EVNN decreases after integrating this information for all exchange rates. The findings can provide assistance to bureaux de change, foreign exchange traders, and governments, enabling them to make well-informed decisions.
No takes yet. Share an insight, caveat, or question.
Alsakarneh et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: