The estimation and interpretation of coefficients in panel gravity models of migration

Abstract

In this paper, we demonstrate that the conventional ordinary least squares and fixed effects estimators of classical gravity models of migration are biased, and that the interpretation of coefficients in the fixed effects gravity model is typically incorrect. We then present a best linear unbiased (BLU) estimator for gravity models of migration, and illustrate its application with interregional data from New Zealand. The results demonstrate that the standard ordinary least squares and fixed effect models lead to biased coefficients on population. Alternative estimates that are BLU are provided for a data generating process with fixed origin and destination effects. The coefficients on population must be interpreted in this model as growth rate effects rather than level effects. Our findings also have significance for other types of spatial interaction modelling.

Empirical Issues in the Homicide-Income Inequality Argument

“…Strain theory has been expanded from the individual to the community level of analysis (Agnew, 1999, 2011; de Beeck et al, 2012; Botchkovar et al, 2018; Brezina et al, 2001; Eitle & Eitle, 2016). This extension of strain theory to the community has been called the macro-level or multi-level version of strain theory (de Beeck et al, 2012; Botchkovar et al, 2018; Brezina et al, 2001). This extension is not surprising as environmental criminological theories frequently extend individual-level criminogenic predictors to communities or other aggregated levels of analysis, given that the criminogenic characteristics of a community are oftentimes the aggregate of the individual criminogenic response to crime (Vilalta, Lopez-Ramirez, et al, 2021).…”

“…Lastly, as measurement error also affects panel data estimators (Nilsson & Agell, 2003), in this study we compare the results of the FD and FE estimates, as attenuation bias varies depending on the method, as explained by Griliches and Hausman (1986). Attenuation bias is less for short term periods but greater for longer periods of time (Cameron & Poot, 2019).…”

Regional migration and the dimension of distance in empirical analysis

“…This modification contributed to the rationalisation of the gravitational migration model. Another application extension of gravity migration models can be panel data for modelling internal and international migration Royuela & Ordóñez, 2018;Cameron & Poot, 2019).…”