Multilevel Modelling with Spatial Interaction Effects with Application to an Emerging Land Market in Beijing, China

Abstract

This paper develops a methodology for extending multilevel modelling to incorporate spatial interaction effects. The motivation is that classic multilevel models are not specifically spatial. Lower level units may be nested into higher level ones based on a geographical hierarchy (or a membership structure—for example, census zones into regions) but the actual locations of the units and the distances between them are not directly considered: what matters is the groupings but not how close together any two units are within those groupings. As a consequence, spatial interaction effects are neither modelled nor measured, confounding group effects (understood as some sort of contextual effect that acts ‘top down’ upon members of a group) with proximity effects (some sort of joint dependency that emerges between neighbours). To deal with this, we incorporate spatial simultaneous autoregressive processes into both the outcome variable and the higher level residuals. To assess the performance of the proposed method and the classic multilevel model, a series of Monte Carlo simulations are conducted. The results show that the proposed method performs well in retrieving the true model parameters whereas the classic multilevel model provides biased and inefficient parameter estimation in the presence of spatial interactions. An important implication of the study is to be cautious of an apparent neighbourhood effect in terms of both its magnitude and statistical significance if spatial interaction effects at a lower level are suspected. Applying the new approach to a two-level land price data set for Beijing, China, we find significant spatial interactions at both the land parcel and district levels.

Geographical scale, industrial diversity, and regional economic stability

“…For example, extending the study region into developing countries like China might provide additional insights into the role of geographical scale issues both in the analytical and the policy aspects of regional economic development. Moreover, as the disaggregated geographical units or lower level units (counties) are often nested within the upper level units (states and EAs), future research should also consider using spatial hierarchical models, such as Dong, Harris, Jones, and Yu () and Lacombe and McIntyre (), to study the scale and scope issues in the diversity–stability relationship.…”

Zero‐inflated multiscale models for aggregated small area health data

“…Third, all the models considered here use the data twice because the data at the PH level are an aggregation of the county-level data. To overcome this issue, multilevel models with spatial interaction effects (Browne, Goldstein, & Rasbash, 2001;Chaix, Merlo, & Chauvin, 2005;Dong, Harris, Jones, & Yu, 2015;Langford, Leyland, Rasbash, & Goldstein, 1999) could be considered.…”

An Inclusive Treadmill? Expansion of Industrial Maize Farming and Simple‐Commodity Producers in Turkey

“…The final variable is a spatially lagged dependent variable that accounts for spatial dependency between provinces (Anselin ; Dong et al ). The spatially lagged variable measures for each province the average level of IMF among the neighboring provinces.…”