Abstract
This paper investigates the short-term dynamics and forecasting of Uzbekistan's monthly product export volume. Given the presence of pronounced trends and seasonal variations in macroeconomic export data, traditional linear models often fail to capture cyclical patterns adequately. To address this, the study applies the Holt-Winters exponential smoothing framework, incorporating both trend and multiplicative seasonal components. The analysis is based on official monthly time series data of export values spanning from January 2020 onward. The raw data is transformed into a formal time series object in RStudio with a fixed seasonal frequency of frequency = 12, reflecting standard monthly periodicity. The findings provide empirical insights into future export trajectories.