旅館營運與地區因素對創造收入影響之研究－－結合階層線性模式與分量迴歸

A Study of the Hotel Operation and Region Factor Effects for Generating Revenue - Combining the Hierarchical Linear Modeling and Quantile Regression

本文係探討旅館營運與地區因素對每人創造收入的影響，應用台灣各地區旅館相關資料，以階層線性模式測定旅館個體與總體變數的多層次影響效果，並進一步應用分量迴歸探討每人創造收入不同分量下，各層次顯著因素的影響性，藉以瞭解其尾端分配資訊。結果顯示：在個體層次部份，除訂房價格與設備支出於觀光旅館外，其餘變數於兩類型旅館中皆具影響性。在總體層次部份，一般旅館的景點數量與道路面積具影響性，人口密度對訂房價格與員工人數具脈絡調節效果；觀光旅館的道路面積具影響性，且對訂房價格具脈絡調節效果；足見地區因素會調節旅館營運因素而影響每人創造收入。此外，由分量迴歸中亦獲致顯著因素對每人創造收入不同分量的影響力不盡相同。

The hierarchical linear model was employed to analyze how revenue generating is related to multi-level influence effects of micro and macro level covariates in the hotel industry. The quantile regression was used to grasp how the covariates affect not only the center of the distribution, but also the spread and the magnitude of extreme events by specifying different covariate effects at different quantile levels for revenue generating. The final results show: At the hotel micro level covariates have a significant effect on revenue generating in tourist and general hotels except for room price and equipment expenditure in tourist hotel. Furthermore, in regional characteristic macro level, the intraclass coefficient shows that the regional differences should not be ignored in both type hotels. Road space and number of attractions have significant effects on revenue generating in general hotels. Furthermore, population density moderates the relationship between room price and revenue generating. Only the road space has significant positive effect on revenue generating in tourist hotels, and this negatively moderates the relationship between room price and revenue generating. Hence, the regional factor impacts revenue generating by moderating operation factors. The results illustrate that the quantile regression models can explore the significant covariates which have varied impacts on revenue generating on different quantiles of the hotel.