Analysis of the Number of International Tourists Visiting Riau Province Over the Last Six Years Using R

This study presents a comprehensive time series analysis of international tourist arrivals to Riau Province, Indonesia, from 2019 to 2024. The research aims to model, compare, and forecast monthly tourist arrivals using the Holt-Winters additive and multiplicative exponential smoothing methods, implemented in R. The dataset, sourced from the official BPS-Statistics Indonesia Riau Province and covering the pre-pandemic, pandemic, and recovery periods, provides a unique opportunity to assess the impact of global disruptions and subsequent recovery on regional tourism. Data preprocessing, model fitting, and evaluation were conducted in R, with model performance assessed using Mean Absolute Error (MAE). The results indicate that the Holt-Winters multiplicative model outperforms the additive model, particularly in capturing the pronounced seasonal fluctuations and level-dependent variance observed in the data. Forecasts for 2025 suggest a continued recovery in international arrivals, with implications for tourism planning and policy in Riau. The study underscores the value of robust time series forecasting for evidence-based decision-making in regional tourism management and highlights the adaptability of R for reproducible, transparent analysis. Recommendations are provided for future research, including the integration of exogenous variables and the exploration of advanced forecasting techniques.