Beyond route-specific forecasting: An empirical test of two cross-series transfer learning strategies for airline demand with short-data constraints

Key Points

• Effective cross-series transfer learning strategies improve airline demand forecasting under short-data constraints.
• The study demonstrates how limited data can still yield accurate predictions in airline demand.
• Using empirical testing, two strategies are evaluated for their forecasting capabilities across different routes.
• Findings highlight the potential for enhanced forecasting methods, indicating the need to adapt to data limitations.