CBN Journal of Applied Statistics (JAS)
Keywords
Exchange rate, harmonization policy, random forest forecasting technique, seasonal autoregressive integrated moving average mode
Abstract
This study employs Random Forest Regression to forecast the Naira/USD exchange rate trajectories after the implementation of the 2023 exchange rate harmonization policy under various scenarios. The forecast was implemented within a policy modeling and evaluation framework that leverages the Seasonal Autoregressive Integrated Moving Average model to predict exchange rate macro-drivers from historical data for the period 2008Q1-2022Q4. The study provides evidence that the harmonization policy would lead to appreciation of the Naira over the first forecast horizon (2023-2026) under specific scenarios. The effectiveness of the policy could further be enhanced when strategically combined with other macroeconomic policies such as reducing inflation and import, promoting economic growth and export, cancelling forex support for the official exchange rate, and completely blocking oil loss. These policies have the potential to further strengthen the Naira in the second forecast horizon (2027-2032), albeit more gradually. The scenarios are specified under the assumptions of non-interference from novel policy shocks such as the removal of petrol subsidy, and absence of other policy-thwarting developments such as poor synchronization or alignment between policies, flinching commitment to the harmonization policy, and market speculations, among others. Hence, the deviation of our exchange rate forecast from the observed reality largely result from violation of our models’ assumptions
Issue
15
Volume
2
First Page
123
Last Page
154
Recommended Citation
Alley, Ibrahim S.
(2024)
"Exchange Rate Harmonization Policy and Naira Exchange Rate Trajectories: A Scenario-Based Modelling for Policy Evaluation,"
CBN Journal of Applied Statistics (JAS): Vol. 15:
No.
2, Article 5.
Available at:
https://dc.cbn.gov.ng/jas/vol15/iss2/5