A method for analyzing the daily variation in the spatial pattern of market area

Abstract

Market area analysis has long been an important research topic in retailing, marketing, and geography. Numerous studies have been conducted in the literature to describe and understand the market area and consumers' behavior. The daily pattern of the market area, however, has not yet been fully analyzed. The market area of department stores and shopping malls is larger on weekends than on weekdays. Many shops and restaurants are closed on Christmas days, which shrinks the market area of shopping malls. To describe and grasp these patterns, this paper proposes a new procedure for analyzing the daily pattern of the market area. Three measures evaluate the difference in the market area between different days and the variation within a day group. A loglikelihood based statistical measure visualizes the spatial difference in the market area between different days. A method for detecting anomalous days on which the market areas are quite different from those of other days. The proposed procedure is applied to the analysis of the visitors of five towns in Tokyo. The results indicate the effectiveness of the procedure as well as provide useful findings for understanding the daily pattern of visitors to the towns.