Measurement Errors Arising When Using Distances in Microeconometric Modelling and the Individuals’ Position Is Geo-Masked for Confidentiality

Econometrics

Giuseppe Arbia²  Giuseppe Espa¹  Diego Giuliani¹ 

[1] Department of Economics and Management, University of Trento, Trento 38122, Italy; E-Mails:;Department of Statistical Science, Catholic University of the Sacred Heart, Rome 00168, Italy

关键词: spatial econometrics;    spatial microeconometrics;    consistency of estimates;    geo-masking;    confidentiality;    distance evaluation;   

DOI  :  10.3390/econometrics3040709

来源: mdpi

【 摘 要 】

In many microeconometric models we use distances. For instance, in modelling the individual behavior in labor economics or in health studies, the distance from a relevant point of interest (such as a hospital or a workplace) is often used as a predictor in a regression framework. However, in order to preserve confidentiality, spatial micro-data are often geo-masked, thus reducing their quality and dramatically distorting the inferential conclusions. In particular in this case, a measurement error is introduced in the independent variable which negatively affects the properties of the estimators. This paper studies these negative effects, discusses their consequences, and suggests possible interpretations and directions to data producers, end users, and practitioners.

【 授权许可】

CC BY   
© 2015 by the authors; licensee MDPI, Basel, Switzerland.