Price Formation and Customers' Choice of Apartments for Short-term Rent in Saint-Petersburg

According to the forecasts of tour operators in 2019, the tourist flow in St. Petersburg is expected to increase from 15 to 20%, which indicates the active development of the Russian hotel market, including online booking. The main goal of the study is to identify and compare the key factors affecting pricing and the probability of apartments booking for daily rent in St. Petersburg. The main difference of this work from others in this area is the application to the same data pricing models from managers as well as from clients of the platform. The research examines 5 groups of factors that influence the price and probability of booking: time,spatial, material, marketing, and characterizing the terms of payment. To assess their impact, 3 types of models are built: a pricing model on the part of managers, a pricing model on the part of clients, and a probability model for a reservation. Each of them is built both on cross sections (hedonistic and logistic regressions) and on the panel (fixed effects model with correction for heteroscedasticity and autocorrelation, Driscoll-Kraay model, panel logit). In particular, the work demonstrated that the pricing in the market of apartments for daily rent has a monthly seasonality. According to the results managers will appreciate the object located near the subway than in the immediate vicinity of attractions, while transport accessibility practically does not determine the price for guests. It may be said that it is desirable to participate in a privileged program of booking.com, because the cost of participation is lower than the potential price premium in terms of visitors. The practical significance of the study is determined by the recommendations developed for the managers of the apartments - on effective management, clients of the platform - on optimal booking, as well as the booking.com platform itself.