Spatial modeling of hotel prices in the Yogyakarta city area using ordinary kriging and cokriging approaches

Abstract

Yogyakarta, renowned as one of Indonesia's most prominent tourist destinations, owes its appeal to its natural beauty, well-preserved environment, and rich cultural and historical heritage. Its reputation as a safe and comfortable destination has led to a consistent annual increase in tourist arrivals. Consequently, there is a growing demand for hotel accommodations that offer competitive pricing to avoid financial losses while meeting tourist expectations. Tourists often rely on ratings and reviews of hotel services and facilities, making these factors significant determinants of pricing strategies. This study aims to provide spatially informed pricing recommendations for potential hotel developments in Yogyakarta using kriging spatial interpolation methods. Two kriging approaches were employed: Ordinary Kriging (OK) and Cokriging (CK), incorporating hotel price and rating data as primary variables. The analysis identified the Exponential semivariogram model as optimal for OK and the Spherical semivariogram model as optimal for CK. Both methods predicted hotel prices around Yogyakarta City to range from IDR 300,000 to IDR 400,000 for locations farther from the city center. Among the two methods, CK demonstrated superior predictive performance, yielding lower Root Mean Square Error (RMSE) and Mean Absolute Percentage Error (MAPE) values compared to OK. These findings highlight the potential of CK for providing accurate and actionable insights into hotel pricing strategies, offering valuable guidance for stakeholders considering investments in Yogyakarta’s thriving hospitality industry.

Keywords: Cokriging, hotel price, interpolation, ordinary kriging
MSC2020: 62H11