共享经济是全球经济新业态、新模式创生的前沿领域之一。在共享经济中，以短期旅行住宿共享为核心内容的民宿在线短租发展令人瞩目，Airbnb则是其中最成功的企业之一，是民宿在线短租平台的鼻祖。2020年年初爆发的新冠肺炎疫情对我国共享经济，尤其是线上短租行业的蓬勃发展带来了巨大的影响。本文期望通过对中国香港地区的Airbnb在线短租房数据进行数据挖掘及建模分析，探究不同房源特征对短租房定价的影响机制，并在此基础上，进一步探究疫情对香港地区在线短租市场的冲击情况。

本文首先建立了普通最小二乘回归和分位数回归模型。回归结果显示，房东所拥有房源总数、房东回复率、最少预订天数和每月评论数量对短租房价格会产生显著的负向影响，而房源所属区划、房间类型、可入住人数、最新评论时间对短租房价格会产生显著的正向影响。较高价位的房源更容易受到地理位置、房间设施和房东个人因素的影响，经营这部分短租房的房东需要注重维持较高的房屋配置，并注重提高服务质量；而较低价位的房源则受到出租时长、评论情况的影响更大，经营这部分短租房的房东更需要保持较高的出租率及好评率。

为了控制模型中可能出现的内生性，本文引入了新冠肺炎疫情冲击作为准自然实验，建立2019年与2020年两年面板数据的双重差分模型。结果显示，新冠肺炎疫情的冲击显著拉低了中心城区短租房源的平均订房价格。疫情发生之后，中心城区房源的平均房价比新界地区低了约110.80元/天。同时，在利用双重差分控制了回归模型中存在的内生性问题后，除了可预订天数变量之外，其余控制变量均对房源价格有显著影响。

The sharing economy is the frontier field of the creation of new global business formats and models. In the sharing economy, the development of online short-term rental of homestays is eye-catching, and Airbnb is one of the most successful companies. The outbreak of the "Covid-19" has had a great impact on the booming development of China’s sharing economy. We aim to explore the impact of different housing characteristics on the prices of Hong Kong's Airbnb online short-term rental housing, and further explore the impact of the COVID-19 epidemic.

First, we established an OLS regression model and a quantile regression model. The results show that in addition to the three variables, the number of days available for booking, the total number of reviews, and the overall rating, the other characteristics will have a significant impact on the price of the listing. Higher-priced listings are more likely to be affected by geographical location, room configuration, and personal factors of the landlord. The landlords operating this part of the listings need to pay more attention to maintaining a higher housing configuration and improving service quality. The lower-priced listings are more affected by the length of rental time and review status. The landlords operating this part of the listings need to maintain a higher occupancy rate and favorable rate.

In order to control the endogeneity that may occur in the model, we introduced the COVID-19 epidemic impact as a quasi-natural experiment and established a DID model. The results showed that the impact of COVID-19 significantly reduced the average price of short-term rental housing in central urban areas. After the outbreak of the epidemic, the average housing price in the central urban area was about 110.80 yuan lower than that in the New Territories. At the same time, after endogeneity is controlled, nearly all the control variables have a significant impact on the housing price.