Accurate exchange rate forecasting is of considerable practical importance for emerging and frontier markets. In such markets, exchange rates tend to exhibit high volatility and may fluctuate significantly due to depreciation pressures, external economic shocks, inflation differentials, and uncertainty in financial markets. This study employs the ARIMA–eGARCH–Monte Carlo modeling framework to forecast the USD/UZS exchange rate. The analysis is based on daily exchange rate data covering the period from January 2018 to April 2026. First, logarithmic exchange rate returns were calculated, and their stationarity was examined using the Augmented Dickey–Fuller (ADF), Phillips–Perron (PP), and KPSS tests. Subsequently, the ARIMA model was employed to capture the conditional mean dynamics of exchange rate returns, while sGARCH, eGARCH, and gjrGARCH models were compared to evaluate volatility clustering and asymmetric shock effects. The empirical results identified the eGARCH(1,1) specification as the best-performing model. In the next stage, 50,000 Monte Carlo simulations were conducted based on the selected model to generate exchange rate forecasts for a 252-trading-day horizon. The simulation results provided probability-based scenarios for the future path of the USD/UZS exchange rate over the following year. The baseline median forecast indicated an exchange rate of UZS 12,518 per US dollar, while the 90% prediction interval ranged from UZS 11,621 to UZS 13,218. The findings suggest that depreciation shocks have a significant impact on the dynamics of the USD/UZS exchange rate and that probability-based forecasting provides a more informative assessment of future exchange rate movements than conventional point forecasts. The proposed approach offers practical implications for the Central Bank, commercial banks, export-import enterprises, and risk managers in assessing foreign exchange risk and conducting stress testing.