Penerapan Model Generalized Space-time Autoregressive (GSTAR) pada Peramalan Tingkat Penghuni Kamar Hotel

Abstract

In the hotel industry, visitors or tourists play a very important role in influencing hotel development. If the number of visitors staying at the hotel is high, then the income earned by the hotel is large. Therefore, a hotel must have a management strategy either through services or through the facilities provided. This is useful to provide satisfaction and attract tourists to return to stay at the hotel. One effort that needs to be done to overcome the problems that occur is to do a forecasting analysis of the Hotel Room Occupancy Rate (TPK). By doing forecasting, the government can find out the number of hotel room occupants that will be received in the future and what needs to be done if there is an increase in hotel room occupants. Currently, forecasting methods are time series or analysistime series much has been done to do forecasting.One of the forecasting methods that can be used in this research is the GSTAR model. Method Generalized Space-Time Autoregressive (GSTAR) is a method used to predict data by considering the relationship between time and location. In this research, the researcher will forecast the Hotel Room Occupancy Rate (TPK) in the provinces of East Java, Central Java, and West Java for the period January 2013 - December 2022 with the aim of knowing the results of the Hotel Room Occupancy Rate forecast in East Java, Central Java, and Java West for the next 5 periods. From the results of the analysis, it was found that the GSTAR model with uniform location weight is the best model. This is because this model meets the assumptions white noise with smaller MSE and MAPE values, namely 34.00442 and 10.28378.