Airbnb房源出租價格之決定因素-特徵價格法之應用

Abstract

隨著網路科技以及行動裝置的革新，促使對等式（Peer to Peer，P2P）網路體系的日漸普及，也讓共享經濟（Sharing Economy）在互聯網中迅速蓬勃的發展，其打破了空間的限制，將各地不同地區閒置之資源進行鏈結與整合。而「Airbnb」也在這股世界浪潮中成長為一間整合閒置房屋擁有者以及租屋需求者的「共享價值鏈」平台，並在短時間內成為了全球三大新創「獨角獸」之一，帶動了閒置房屋於網頁平台出租的風潮與發展。 隨著網路平台以及資訊科技的日漸完善，出租房源的資訊於網頁上越來越詳細，消費者所面對到的房源選擇不再單以房源描述特徵為出租房源的選擇，除了基本的房源特徵資訊，也新增了許多旅客對於該房源的體驗評分做為該房源之特徵描述，因此，也帶給了旅客決定住宿過程中，增添了許多與以往不一樣的選擇條件。 本研究應用自動化程式爬蟲軟體，並加入特徵價格法為應用，蒐集「臺灣Airbnb」平台上之出租房源資訊以及交易時空下的社會經濟變數，並且應用普通最小平方法模型分析影響Airbnb房源出租之因素，實證結果發現，不同的房源評價分數以及旅館業、觀光業競爭對手，對於「Airbnb」的交易出租房源有顯著的影響，並且進一步以分量迴歸模型加以分析，將樣本分為合住房間、獨立房間以及整棟房屋三種房源，結果亦有顯著的不同，顯示出租房源可被分成不同類型，且每個類型於不同分量下存在不同的影響效果。希望能透過本研究結果幫助有益於網頁平台出租的房東以及Airbnb平台業者提出建議，並且為共享經濟於臺灣之發展貢獻相關的研究之參考。

With the innovation of network technology and mobile devices, Peer to Peer (P2P) network system has become more popular than before, and the 'Sharing Economy' has developed dramatically in this Internet wave, which has broken the Restrictions of space. Linking and integrating idle resources in different regions. 'Airbnb' has also grown with this International wave, which built a 'shared value chain' platform that integrates idle homeowners and renters, and has become the world's three new 'unicorns' in a short period of time. It has driven the wave of development and rental of idle houses on the web platform. With the improvement of the Internet platform and information technology, the information is much clear and plentiful on the online rental platform. Housing rental not only described basic information but also described a lot of users' experience and the property of rating score. Therefore, choice the rental house will have more different condition to understand the target and its feature. The data of this paper are collected from 'Taiwan Airbnb' platform, with automated program crawler software, and followed the Hedonic Pricing Method to describe the independence variables, which including rental listing information on the 'Taiwan Airbnb' platform, crime rate, traffic accident rate, competitor and socio-economic variables. This paper divided the data into three types of housing style, which is private room, entire apartment and shared room, and used ordinary least squares model and Quantile regression model to analyze the 'Effect of Airbnb pricing'. We hope that the result of this research will support landlords and 'Taiwan Airbnb' can make appropriate strategy. Moreover, I hope to contribute the references of sharing economic in Taiwan.