抄録
This study proposes a stochastic model of the closure of grocery stores using spatio-temporal data. The model is based on the partial logistic model where customers follow a probabilistic store choice model. The model allows us not only to describe the past closure of grocery stores but also to consider the potential of their closure in future.
To test the validity of the model, we applied it to the analysis of the grocery stores from 2000 to 2005 in Gunma, Tochigi, and Ibaraki prefectures. The result showed that the model can predict the closure of grocery stores in these areas with high accuracy, especially small grocery stores located in the regions whose density of stores is low.
