Forecasting of Hotel Revenue, Bookings and Cancellations through Machine Learning and Deep Learning
Romanian Journal of Economic Forecasting, cilt.29, sa.1, ss.105-125, 2026 (SSCI, Scopus)
- Yayın Türü: Makale / Tam Makale
- Cilt numarası: 29 Sayı: 1
- Basım Tarihi: 2026
- Dergi Adı: Romanian Journal of Economic Forecasting
- Derginin Tarandığı İndeksler: Social Sciences Citation Index (SSCI), Scopus, EconLit
- Sayfa Sayıları: ss.105-125
- Anahtar Kelimeler: deep learning, machine learning, tourism industry
- Sivas Cumhuriyet Üniversitesi Adresli: Evet
Özet
For managers, making the right decisions in a dynamic environment becomes critical. This study explores how forecasting methods convert reservation data into actionable information for decision-making in the hospitality industry. Based on data obtained from 5,087 accommodation facilities in Türkiye, this study forecast reservations, number of nights, cancellations, and revenue using deep learning model (LSTM) and machine learning models (linear regression, robust linear regression, artificial neural networks). Comparative analyses reveal that LSTM achieves the highest forecast accuracy and offers a valuable tool for managerial decision-making. This finding demonstrates the practical application of advanced forecasting methods, enabling managers to make more accurate decisions in dynamic and uncertain environments. The study also examines this issue using real data.