Combining microsimulation and spatial interaction models for retail location analysis |
| |
| Authors: | Tomoki Nakaya A Stewart Fotheringham Kazumasa Hanaoka Graham Clarke Dimitris Ballas Keiji Yano |
| |
| Institution: | (1) Department of Geography, Ritsumeikan University, Kyoto, Japan;(2) National Centre for Geocomputation, John Hume Building, National University of Ireland-Maynooth, Maynooth, Kildare, Ireland;(3) School of Geography, University of Leeds, Leeds, LS2 9JT, UK;(4) Department of Geography, University of Sheffield, Winter Street, Sheffield, S10 2TN, UK |
| |
| Abstract: | Although the disaggregation of consumers is crucial in understanding the fragmented markets that are dominant in many developed
countries, it is not always straightforward to carry out such disaggregation within conventional retail modelling frameworks
due to the limitations of data. In particular, consumer grouping based on sampled data is not assured to link with the other
statistics that are vital in estimating sampling biases and missing variables in the sampling survey. To overcome this difficulty,
we propose a useful combination of spatial interaction modelling and microsimulation approaches for the reliable estimation
of retail interactions based on a sample survey of consumer behaviour being linked with other areal statistics. We demonstrate
this approach by building an operational retail interaction model to estimate expenditure flows from households to retail
stores in a local city in Japan, Kusatsu City.
|
| |
| Keywords: | Spatial interaction modelling Microsimulation Retail interaction Consumer survey Market segmentation |
| 本文献已被 SpringerLink 等数据库收录！ |
|
