This study proposes a hotel demand estimation mechanism that assesses the likelihood of forthcoming occupancy peaks and troughs applicable to different hotel classifications. In anticipating rate fluctuations, the approach is less dependent than many prevailing hotel forecasts on short-term seasonal-related factors. In operating revenue management systems, hotel managers should predict forthcoming occupancy upturns and downturns to prepare accurate midto long-run estimates. The proposed approach reduces the financial risks associated with volatile occupancy rates and facilitates efficient resource management. The average contraction period for Hong Kong hotel occupancies from one peak point to the next trough was found to exceed the duration of the corresponding expansion period.