Using Web-Data to Estimate Spatial Regression Models

Macro econometrics has been recently affected by the so-called ‘Google Econometrics’. Comparatively less attention has been paid to the subject by the regional and spatial sciences where the Big Data revolution is challenging the conventional econometric techniques with the availability of a variety...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:International regional science review Jg. 47; H. 2; S. 204 - 226
Hauptverfasser: Arbia, Giuseppe, Nardelli, Vincenzo
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Los Angeles, CA SAGE Publications 01.03.2024
Sage Publications Ltd
Schlagworte:
Bayesian analysis
Big Data
Crowdsourcing
Econometrics
Inference
Spatial analysis
Statistics
spatial microeconometrics
spatial regression
webscraping
crowdsourcing
big data
ISSN:0160-0176, 1552-6925
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Macro econometrics has been recently affected by the so-called ‘Google Econometrics’. Comparatively less attention has been paid to the subject by the regional and spatial sciences where the Big Data revolution is challenging the conventional econometric techniques with the availability of a variety of non- traditionally collected data (such as, e. g., crowdsourcing, web scraping, etc) which are almost invariably geo-coded. However, these unconventionally collected data represent only what in statistics is known as a “convenience sample” that does not allow any sound probabilistic inference. This paper aims at making aware researchers of the consequence of the unwise use of such data in the applied work and to propose a technique to minimize such the negative effects in the estimation of spatial regression. The method consists of manipulating the data prior their use in an inferential context.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0160-0176
1552-6925
DOI:10.1177/01600176231173438