Time Series Analysis of Exchange Rate Inflation and Interest Rate on Growth in Pakistan

Abstract

This study examines the dynamic linkages among real GDP, exchange rate, inflation rate, and real interest rate in Pakistan from 2000 to 2022 using World Bank data. A Vector Autoregression (VAR) model was applied after confirming stationarity with the Augmented Dickey-Fuller (ADF) test, and diagnostic checks confirmed model adequacy. The analysis, conducted in R and EViews, reveals significant interdependencies among the variables, indicating that past values hold predictive power for future outcomes. Impulse Response Functions (IRFs) highlight that real GDP responds positively to shocks in its own values and in the real interest rate, whereas the exchange rate and inflation exhibit notable fluctuations. Forecast accuracy was evaluated using RMSE, MAE, MAPE, and Theil’s U, with results showing stronger predictive performance for real GDP compared to exchange rate and inflation. These findings emphasize the relevance of macroeconomic interactions for policy formulation, while also suggesting the need for methodological refinement to improve forecast precision, particularly for the exchange rate and inflation.