Predicting Ship Prices with Smart Tech

The shipping industry is always looking for better ways to value ships. This is key for buying, selling, and insuring them. One modern approach is using Machine Learning (ML). This method has shown it can outperform old-school techniques. The goal here is to create a precise ship valuation model for supramax and ultramax ships. These are medium to large cargo ships. The process involved four main steps. The first step was to gather data. This data covered ship sales from August 2005 to December 2022. The focus was on finding important factors that affect ship prices. These factors were then used to build linear regression models. Two models were created. The first used the Baltic Exchange indices from the month of sale. The second used indices from the month of sale and previous months. Next, these models were tested using different ML algorithms. Linear Regression, Decision Tree, Random Forest, and XGBoost were used to predict prices. The idea was to see which model and algorithm combo worked best. The results showed that the models were quite accurate. This was a good sign that ML could be a reliable tool for ship valuation. The third step was to simplify the models. The goal was to use factors that are easy to get from the market. This made the models more practical. The simplified models were then re-tested with the same ML algorithms. The results were close to the original models. This showed that the simplified models could also be used for predictions. Finally, the models were tested with new data from 2023. This was to see if they could handle more information. The results were even better. The model that used Baltic Exchange indices from previous months performed the best. XGBoost was found to be the most effective algorithm. This whole process shows that ML can be a powerful tool for predicting ship prices. It can handle complex data and provide accurate results. This is important for the shipping industry. It helps in making better decisions when buying, selling, or insuring ships.