earticle

영어

This study focuses on forecasting maritime freight rates by applying a time-series decomposition approach to two key indicators: the ClarkSea Index, which represents overall shipping market conditions, and Clarksons Average Tanker Earnings, which measure vessel-specific profitability. The analysis utilizes monthly data from October 2019 to September 2024, sourced from Clarksons Research via the Shipping Intelligence Network. By leveraging 60 months of historical data, the study generates one-month, two-month, and three-month ahead forecasts and evaluates their accuracy against actual observed values. As an analytical approach, the time series data are decomposed into three key components: trend-cycle, seasonal variation, and irregular components, and forecasting evaluation is conducted based on this decomposition. To assess the accuracy of the forecasting model, the Mean Absolute Percentage Error is utilized. The results indicate that forecasting errors remain below 10%, demonstrating a high degree of reliability. Specifically, the model performs well in predicting Clarksons Average Tanker Earnings, with minimal deviations from actual values.