Modeling of EUR-MAD and USD-MAD Exchange Rates
Currency Risk can be very problematic for our companies, so it is important for Moroccan companies exposed to currency risk to be able to measure this exposure in order to manage it properly.
The objective of this paper is to estimate the foreign exchange risk EUR / MAD and USD / MAD according to Basel II recommendations on capital requirements for market risks. To quantify the potential capital loss on foreign exchange risk, the experts use the VAR (Value At Risk) model to determine the potential loss that could impact the firm's performance.
We applied the VAR concept to estimate the foreign exchange risk of the Moroccan Dirham against the Euro and then against the US Dollar. Three standard VAR calculation models most used by risk managers (historical method, parametric method and Monte Carlo simulation) were studied and then applied to a historical exchange rate EUR / MAD and USD / MAD over the period from 02/01/2017 to 31/12/2018 and on a database of 520 observations.
Our study revealed that the Monte Carlo simulation method recorded the highest value of the VaR which is established at 281.170 DH against the parametric method which is limited to 204.100 DH. The average of the three methods is around 246.170 DH which represents a loss percentage of 0.91% of the total value of the portfolio.
However, to complete this study, we believe that it is wise to use stress testing and backtesting tests which make it possible to validate this VaR model and to meet the standards set by the Basel Committee.
With a confidence level of 99%, we found three exceedances of the losses compared to the values of the average VaR. For a 95% confidence level, we have 16 exceedances. The maximum loss corresponds to the date of: 24/08/2017, explained by the sharp drop in the EUR / MAD rate, from 11.235 to 11.156 DH.
Copyright (c) 2021 Ahmed Hrifa, Zineb Bamousse
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
All articles published in this journal are licenced uder a creative commons attribution-noncommercial-noderivatives 4.0 international licence