Consider a situation when a customer looks for the optimal price versus quality combination to book a hotel room on the Internet. There are numerous online distribution channels (intermediaries) where a person can order a room with desired characteristics. The task of this paper is to develop certain pricing models that would be used to set an optimal hotel room price for any intermediary and for any type of stay. Thus, the central question is: by what means can a lodging company react swiftly to the changing situation at the online distributer and set the prices accordingly? This problem is important and relevant for two reasons. First, my associate company is developing product Y that will have a function to manage online prices in the hospitality industry. Second, this topic relates to the revenue management discipline that is currently on the verge of the paradigm shift from the long- to the short-term pricing. Achieved results suggest the particular pricing methods for how lodging services can be evaluated on the short-term basis. Econometric techniques were a fundamental tool for the empirical analysis. I have used the ordinary least squares (OLS) method to estimate the parameters in the linear regression models. Two types of data were at disposal for this purpose: prospective information for several hotels from an online intermediary Booking.com and historical information provided by two hoteliers. The key finding that shaped the final results and basically the whole structure of the paper was the simultaneity bias problem. I have chosen and tested two ways to deal with this issue: the instrumental variables with the two-stage least squares method and the autoregressive integrated moving average (ARIMA) technique. The ultimate results put forward two pricing options. First is the pooled cross-section data approach where various explanatory variables are used and instruments are applied to control for the endogeneity. Second is the time series method that has only the lagged values of the regressand as independent variables. The first way is explanatory and the second one – descriptive. The implications of these results are straightforward: developed pricing methods can be used as a function of the particular product Y and they give the specific ideas for scholars in the revenue management science on how online prices might be determined.
|Educations||MSc in Applied Economics and Finance, (Graduate Programme) Final Thesis|
|Number of pages||97|