Modelling and forecast of Uganda’s merchandise trade balance: application of ardl and arima-garch models

dc.contributor.author Niwagaba. Judith. Stella
dc.date.accessioned 2025-11-28T16:15:42Z
dc.date.available 2025-11-28T16:15:42Z
dc.date.issued 2025-11
dc.description A dissertation submitted to the Directorate of Research and Graduate Training in partial fulfilment of the requirements for the award of a Degree of Master of Statistics of Makerere University.
dc.description.abstract Uganda’s merchandise trade balance has persistently remained in deficit, posing challenges to external sector stability and sustainable economic growth. This study modelled and forecasted Uganda’s merchandise trade balance using the Autoregressive Distributed Lag (ARDL) and ARIMA–GARCH frameworks to identify macroeconomic determinants and predict volatility. Monthly time-series data from July 2001 to June 2024 were obtained from the Bank of Uganda and the World Development Indicators. The ARDL approach was applied to estimate both short- and long-run determinants of the trade balance, while the ARIMA–GARCH model was employed to analyse volatility clustering and forecast future trade balance movements. The findings show that in the long run, the export price index, foreign direct investment, and GDP significantly influence Uganda’s trade balance. In particular, EPI and FDI positively affect trade performance, whereas the import price index exerts a negative effect, suggesting that higher import costs worsen trade deficits. In the short run, lagged trade balance, GDP growth, and price indices significantly determine trade outcomes, Specifically, current GDP growth negatively impacts the trade balance (D(GDP) = -0.3735, while its lagged value shows a positive effect (D(GDP) [-1] = 0.2999. EPI and IPI positively influence short-run trade performance, with coefficients of 0.5160 and, respectively. with the error-correction term of -0.5686 confirming strong convergence toward equilibrium. The GARCH model results indicated substantial volatility clustering in Uganda’s trade balance, implying sensitivity to external shocks and policy changes. Forecast simulations from the ARIMA–GARCH model projected persistent trade deficits through 2027, with cyclical fluctuations peaking around mid-2026. The study concludes that Uganda’s trade imbalance is both structural and cyclical, driven by external price dynamics and domestic macroeconomic factors. It recommends policies that promote export value addition, diversification, and FDI in export-oriented sectors, alongside effective risk-management frameworks and early-warning systems. Strengthening macroeconomic fundamentals, improving competitiveness, and reducing import dependency will be essential for achieving a more balanced and sustainable trade position in the long term. Subject keywords; Uganda’s merchandise trade balance
dc.identifier.citation Niwagaba, J. S. Modelling and forecast of Uganda’s merchandise trade balance: application of ardl and arima-garch models. Unpublished master’s thesis, Makerere University
dc.identifier.uri https://makir.mak.ac.ug/handle/10570/15366
dc.language.iso en
dc.publisher Makerere University
dc.title Modelling and forecast of Uganda’s merchandise trade balance: application of ardl and arima-garch models
dc.type Other
Files
Original bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
Niwagaba-cobams-masters-2025.pdf
Size:
1.73 MB
Format:
Adobe Portable Document Format
Description:
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
1.71 KB
Format:
Item-specific license agreed upon to submission
Description: