基于二项logit模型的房型选择行为分析

Analysis of Room Type Selection Behavior Based on Binomial Logit Model

目前,学者多利用集计模型分析、预测酒店顾客群体的客房需求,忽略了个体间的差异.针对不同房型分别建立非集计的二项logit模型,分析顾客的不同属性:性别、住宿天数、入住时间、退房时间和月份对其房型的选择行为的影响,并预测不同房型的预定概率.结果表明:商务/行政大床房、商务/行政双床房、商务/行政三人套房在不同因素的影响下,其预定概率存在显著差异.为酒店合理应对波动较大的顾客需求提供了理论基础.

Traditional method of analyzing and predicting room demand of different customer groups in the hotel industry was using aggregate models,which ignored the differences between individuals.In this paper,we established Binary Logit Models,which are a type of disaggregate model,to analyze the influences of different attributes,including customer’s gender,length of stay,check-in time,check-out time and month,on their roomtype choice behaviors,then we predicted the predetermined probability of different room types.The results indicated that there are significant differences in the predetermined probability of business/administrative queen room,business/administrative twin room,and business/administrative three-person suite under different attributes.This paper provides a theoretical basis for the hotel industry to respond appropriately to the volatility of customer needs.