Résumés
Résumé
Les modèles d’économétrie spatiale sont de plus en plus utilisés en géographie. Récemment, des extensions des modèles spatiaux autorégressifs ont été proposées afin d’analyser des données par panel, comprenant N entités spatiales pour lesquelles un ensemble de variables sont observées à T dates. À l’heure actuelle, ces modèles de régression par panel sont très peu connus des géographes. Par conséquent, notre objectif, dans cet article, est de décrire ces modèles et d’illustrer leurs avantages pour l’élaboration de diagnostics urbains longitudinaux. Pour ce faire, nous proposons un retour sur les régressions par panel. Puis nous décrivons une routine méthodologique permettant de déterminer le modèle le plus approprié au jeu de données à l’étude, parmi plusieurs modèles de régression par panel. L’application empirique, portant sur la modélisation de la pauvreté à Montréal de 1986 à 2016 par secteur de recensement, démontre que cette méthode est très robuste pour contrôler la variable de dépendance spatiale, mais aussi pour établir plusieurs types d’effets (directs, de débordement et de renvoi).
Mots-clés :
- Pauvreté urbaine,
- analyse spatiale,
- régression spatiale,
- régression par panel,
- Montréal
Abstract
There is a growing interest for spatial econometric models in geography. Recently, extensions of the cross-sectional spatial autoregressive models have been proposed to consider the temporal dimension of panel data (N locations at T time periods). Currently, these models are still little known to geographers. Therefore, the aim of this paper is to describe these models and present their advantages in analyzing the linear relation between a variable of interest and explanatory variables in the case of urban panel data. We first summarize panel models and their specifications. Then, we describe a methodological routine to identify the most appropriate model for a spatial panel dataset among various spatial panel models. Our empirical application of this method on poverty in the region of Montreal (at census tract level) from 1986 to 2016 shows its robustness to control for spatial dependency, but also its ability to identify several spatial effects ignored by classical aspatial models (spillover effects of the dependent and independent variables and feedback effect).
Keywords:
- Urban poverty,
- spatial analysis,
- spatial regression,
- spatial panel regression,
- Montreal
Resumen
En Geografía, los modelos de econometría espacial son utilizados con mayor frecuencia. Últimamente, extensiones de modelos espaciales auto regresivos han sido propuestos para analizar datos por panel (muestra de grupo permanente) con N entidades espaciales, por las que se observan un conjunto de variables en T data. Actualmente, los modelos de regresión por panel son poco conocidos por los geógrafos. Nuestro objetivo, en el presente artículo, es de describirlos e ilustrar sus ventajas para la elaboración de diagnósticos urbanos longitudinales. Para ello, proponemos una retrospectiva sobre las regresiones por panel. Luego, describimos una rutina metodológica que permita determinar el modelo apropiado a la serie de datos estudiados, entre los varios modelos de regresión por panel. La aplicación empírica, que concierne la modelización de la pobreza en Montreal de 1986 a 2016 por sector de empadronamiento, demuestra la robustez del método, tanto para controlar la variable de dependencia espacial, como para establecer varios tipos de efectos (directos, de exceso y de exclusión).
Palabras clave:
- Pobreza urbana,
- análisis espacial,
- regresión espacial,
- regresión por panel,
- Montreal
Parties annexes
Bibliographie
- ADES, Josefina, APPARICIO, Philippe et SÉGUIN, Anne Marie (2012) Are new patterns of low income distribution emerging in Canadian metropolitan areas? The Canadian Geographer, vol. 56, no 3, p. 339-361.
- ANSELIN, Luc (1988) Spatial econometrics: Methods and models. Dordrecht, Kluwer Academic Publishers.
- APPARICIO, Philippe, CARRIER, Mathieu, GELB, Jérémy, SÉGUIN, Anne-Marie et KINGHAM, Simon (2016) Cyclists’ exposure to air pollution and road traffic noise in central city neighbourhoods of Montreal. Journal of Transport Geography, vol. 57, p. 63-69.
- APPARICIO, Philippe, SÉGUIN, Anne-Marie et LELOUP, Xavier (2007) Modélisation spatiale de la pauvreté à Montréal : apport méthodologique de la régression géographiquement pondérée. Le Géographe canadien, vol. 51, no 4, p. 412-427.
- BALTAGI, Badi H. (2013) Econometric analysis of panel data. Hoboken, John Wiley and Sons, Inc.
- BIVAND, Roger, PEBESMA, Edzer J. et GÓMEZ-RUBIO, Virgilio (2013) Applied spatial data analysis with R. New York, Springer.
- BOUAYAD-AGHA, Salima, LE GALLO, Julie et VEDRINE, Lionel (2018) Économétrie spatiale sur données de panel. Dans Vincent Loonis (dir.) Manuel d’analyse spatiale. Théorie et mise en oeuvre pratique avec R. Paris, Institut national de la statistique et des études économiques, p. 183-209.
- BURNETT, J. Wesley, BERGSTROM, John C. et DORFMAN, Jeffrey H. (2013) A spatial panel data approach to estimating US state-level energy emissions. Energy Economics, vol. 40, no C, p. 396-404.
- CAN, Ayse (1996) Weight matrices and spatial autocorrelation statistics using a topological vector data model. InternationalJournal of Geographical Information Systems, vol. 10, no 10, p. 1009-1017.
- CARRIER, Mathieu, APPARICIO, Philippe, SÉGUIN, Anne‑Marie et CROUSE, Dan (2014) The application of three methods to measure the statistical association between different social groups and the concentration of air pollutants in Montreal: A case of environmental equity. Transportation Research Part D: Transport and Environment, vol. 30, p. 38-52.
- CARRIER, Mathieu, APPARICIO, Philippe et SÉGUIN, Anne‑Marie (2016a) Road traffic noise in Montreal and environmental equity: What is the situation for the most vulnerable population groups? Journal of Transport Geography, vol. 51, p. 1-8.
- CARRIER, Mathieu, APPARICIO, Philippe et SÉGUIN, Anne‑Marie (2016b) Road traffic noise geography during the night in Montreal: An environmental equity assessment. The Canadian Geographer, vol. 60, no 3, p. 394-405.
- CARRIER, Mathieu, APPARICIO, Philippe, KESTENS, Yan, SÉGUIN, Anne-Marie, PHAM, Hien et al. (2016c) Application of a Global Environmental Equity Index in Montreal: Diagnostic and further implications. Annals of the American Association of Geographers, vol. 106, no 6, p. 1268-1285.
- CHATTERJEE, Samprit et HADI, Ali (2015) Regression analysis by example. Hoboken, John Wiley & Sons, Inc.
- CRESSIE, Noel (2015) Statistics for spatial data. Hoboken, John Wiley & Sons, Inc, [1993].
- CROISSANT, Yves et MILLO, Giovanni (2008) Panel data econometrics in R: The plm package. Journal of Statistical Software, vol. 27, no 2, p. 1-43.
- DEVAUX, Nicolas, DUBÉ, Jean et APPARICIO, Philippe (2017) Anticipation and post-construction impact of a metro extension on residential values: The case of Laval (Canada), 1995–2013. Journal of Transport Geography, vol. 62, p. 8-19.
- DORMANN, Carsten, MCPHERSON, Jana, ARAÚJO, Miguel, BIVAND, Roger, BOLLIGER, Janine et al. (2007) Methods to account for spatial autocorrelation in the analysis of species distributional data : A review. Ecography, vol. 30, no 5, p. 609-628.
- DUBÉ, Jean et LEGROS, Diègo (2014a) Spatial econometrics and the hedonic pricing model: What about the temporal dimension? Journal of Property Research, vol. 31, no 4, p. 333-359.
- DUBÉ, Jean et LEGROS, Diègo (2014b) Économétrie spatiale appliquée des microdonnées. Londres, ISTE Group.
- DUBÉ, Jean, LEGROS, Diègo, THÉRIAULT, Marius et DES ROSIERS, François (2014) A spatial Difference-in-Differences estimator to evaluate the effect of change in public mass transit systems on house prices. Transportation Research Part B: Methodological, vol. 64, p. 24-40.
- DUBÉ, Jean et POLÈSE, Mario (2015) The view from a lucky country: Explaining the localised unemployment impacts of the Great Recession in Canada. Cambridge Journal of Regions, Economy and Society, vol. 9, no 1, p. 235-253.
- DUBIN, Robin (2008) Spatial weights. Dans A. Stewart Fotheringham et Peter A. Rogerson (dir.) The SAGE handbook of spatial analysis. Thousand Oaks, SAGE Publications, p. 125-158.
- DUVIVIER, Chloé, POLÈSE, Mario et APPARICIO, Philippe (2018) The location of information technology-led new economy jobs in cities: Office parks or cool neighbourhoods? Regional Studies, vol. 52, no 6, p. 756-767.
- ELHORST, J. Paul (2009) Spatial panel data models. Dans Manfred M. Fischer et Arthur Getis (dir.) Handbook of applied spatial analysis: Software tools, methods and applications. Berlin, Springer Science & Business Media, p. 377-407.
- ELHORST, J. Paul (2010) Applied spatial econometrics: Raising the bar. Spatial Economic Analysis, vol. 5, no 1, p. 9-28.
- ELHORST, J. Paul (2012) Matlab software for spatial panels. International Regional Science Review, vol. 37, no 3, p. 389-405.
- ELHORST, J. Paul (2014) Spatial econometrics: From cross-sectional data to spatial panels. New York, Springer Publishing.
- JELINSKI, Dennis E. et WU, Jianguo (1996) The modifiable areal unit problem and implications for landscape ecology. Landscape Ecology, vol. 11, no 3, p. 129-140.
- FOTHERINGHAM, Stewart, BRUNSDON, Chris et CHARLTON, Martin (2003) Geographically weighted regression: The analysis of spatially varying relationships. Hoboken, John Wiley & Sons, Inc.
- GE, Xiangyu, ZHOU, Zhimin, ZHOU, Yanli, YE, Xinyue et LIU, Songlin (2018) A spatial panel data analysis of economic growth, urbanization, and NOx emissions in China. International Journal of Environmental Research and Public Health, vol. 15, no 4, p. 725-745.
- GOUVERNEMENT DU CANADA (2016) Document d’information sur la pauvreté au Canada. Ottawa, Gouvernement du Canada.
- GRIFFITH, Daniel (2013) Spatial autocorrelation and spatial filtering: Gaining understanding through theory and scientific visualization. Berlin, Springer Science & Business Media.
- GRIFFITH, Daniel et PERES-NETO, Pedro (2006) Spatial modeling in ecology: The flexibility of eigenfunction spatial analyses. Ecology, vol. 87, no 10, p. 2603-2613.
- GROULX, Lionel-Henri (2011) Les facteurs engendrant l’exclusion au Canada : survol de la littérature multidisciplinaire. Québec, Centre d’étude sur la pauvreté et l’exclusion.
- HAINING, Robert (2003) Spatial data analysis: Theory and practice. Cambridge (Angleterre), Cambridge University Press.
- HAUSMAN, Jerry, HALL, Bronwyn H. et GRILICHES, Zvi (1984) Econometric models for count data with an application to the patents-R&D relationship. Technical working paper, no 17, Cambridge (États-Unis), National Bureau of Economic Research.
- HEISZ, Andrew et MCLEOD, Logan (2004) Faible revenu dans les régions métropolitaines de recensement, 1980 à 2000. Ottawa, Statistique Canada.
- HUANG, Qiong et CHAND, Satish (2015) Spatial spillovers of regional wages: Evidence from Chinese provinces. China Economic Review, vol. 32, p. 97-109.
- HUANG, Jianhuan et XIA, Jiejin (2016) Regional competition, heterogeneous factors and pollution intensity in China: A spatial econometric analysis. Sustainability, vol. 8, no 171, p. 1-26.
- KEITT, Timothy H., BJØRNSTAD, Ottar, DIXON, Philip M. et CITRON POUSTY, Steve (2002) Accounting for spatial pattern when modeling organism environment interactions. Ecography, vol. 25, no 5, p. 616-625.
- KESTENS, Yan, THÉRIAULT, Marius et DES ROSIERS, François (2006) Heterogeneity in hedonic modelling of house prices: Looking at buyers’ household profiles. Journal of Geographical Systems, vol. 8, no 1, p. 61-96.
- KOPCZEWSKA, Katarzyna, KUDŁA, Janusz et WALCZYK, Konrad (2017) Strategy of spatial panel estimation: Spatial spillovers between taxation and economic growth. Applied Spatial Analysis and Policy, vol. 10, no 1, p. 77-102.
- LEE, Kevin Kaalip (2000) Urban poverty in Canada: A statistical profile. Ottawa, Canadian Council on Social.
- LELOUP, Xavier (2007) Vers la ville pluraliste ? Distribution et localisation des minorités visibles à Montréal, Toronto et Vancouver en 2001. Canadian Journal of Regional Science, vol. 30, no 2, p. 263-292.
- LESAGE, James (2014) Spatial econometric panel data model specification: A Bayesian approach. Spatial Statistics, vol. 9, p. 122-145.
- LESAGE, James et PACE, Kelly (2009) Introduction to spatial econometrics. Boca Raton, Chemical Rubber Company Press.
- LIU, Yu, XIAO, Hongwei et ZHANG, Ning (2016) Industrial carbon emissions of China’s regions: A spatial econometric analysis. Sustainability, vol. 8, no 3, p. 210-224.
- MILLO, Giovanni et PIRAS, Gianfranco (2012) splm: Spatial panel data models in R. Journal of Statistical Software, vol. 47, no 1, p. 1-38.
- OPENSHAW, Stan (1977) Optimal zoning systems for spatial interaction models. Environment and planningA, vol. 9, no 2, p. 169-184.
- PARAJULI, Jitendra et HAYNES, Kingsley (2017) Panel data models of new firm formation in New England. Region: The Journal of ERSA, vol. 4, no 3, p. 65-76.
- PFARR, Christian, SCHMID, Andreas et SCHNEIDER, Udo (2010) Estimating ordered categorical variables using panel data: A generalized ordered probit model with an autofit procedure. Paper no 24181, Munich, Munich Personal RePEc Archive.
- PHAM, Thi Thanh Hiên, APPARICIO, Philippe, SÉGUIN, Anne-Marie, LANDRY, Shawn et GAGNON, Martin (2012) Spatial distribution of vegetation in Montreal: An uneven distribution or environmental inequity? Landscape and Urban Planning, vol. 107, no 3, p. 214-224.
- PHAM, Thi Thanh Hiên, APPARICIO, Philippe, LANDRY, Shawn, SÉGUIN, Anne-Marie et GAGNON, Martin (2013) Predictors of the distribution of street and backyard vegetation in Montreal, Canada. Urban forestry & Urban Greening, vol. 12, no 1, p. 18-27.
- PLUMMER, Lawrence A. (2009) Spatial dependence in entrepreneurship research: Challenges and methods. Organizational Research Methods, vol. 13, no 1, p. 146-175.
- QIAN, Haifeng et ZHAO, Chang (2018) Space-time analysis of high technology entrepreneurship: A comparison of California and New England. Applied Geography, vol. 95, no 1, p. 111-119.
- RUE, Havard et HELD, Leonhard (2005) Gaussian Markov random fields: Theory and applications. Londres, Chapman and Hall/CRC Press.
- SÉGUIN, Anne-Marie, APPARICIO, Philippe et RIVA, Mylène (2012) Identifying, mapping and modelling trajectories of poverty at the neighbourhood level: The case of Montréal, 1986–2006. Applied Geography, vol. 35, nos 1-2, p. 265-274.
- SÉGUIN, Anne-Marie, APPARICIO, Philippe, RIVA, Mylène et NEGRON-POBLETE, Paula (2016) The changing spatial distribution of Montreal seniors at the neighbourhood level: A trajectory analysis. Housing Studies, vol. 31, no 1, p. 61-80.
- STATISTIQUE CANADA (2010) Recensement de la population, 1986. No 97-550-XWF2006001 au catalogue, Ottawa, Gouvernement du Canada.
- STATISTIQUE CANADA (2018) Dictionnaire. Recensement de la population, 2016. No 98-301-X2016001 au catalogue, Ottawa, Gouvernement du Canada.
- STUDENMUND, Arnold et CASSIDY, Henry (2016) Using econometrics: A practical guide. Londres, Pearson.
- R CORE TEAM (2017) R:A language and environment for statistical computing. Vienne, R Foundation for Statistical Computing.
- THANOS, Sotirios, DUBÉ, Jean et LEGROS, Diègo (2016) Putting time into space: The temporal coherence of spatial applications in the housing market. Regional Science and Urban Economics, vol. 58, p. 78-88.
- TONG, Tingting, YU, Tun-Hsiang Edward, CHO, Seong-Hoon, JENSEN, Kimberly et DE LA TORRE UGARTE, Daniel (2013) Evaluating the spatial spillover effects of transportation infrastructure on agricultural output across the United States. Journal of Transport Geography, vol. 30, p. 47-55.
- WONG, David et AMRHEIN, Carl (1996) Research on the MAUP: Old wine in a new bottle or real breakthrough?Journal of Geographical Systems, vol. 3, no 2, p. 73-76.
- WOOD, Simon N. (2017) Generalized additive models: An introduction with R. Londres, Chapman and Hall/CRC.
- WOOLDRIDGE, Jeffrey M. (2002) Econometric analysis of cross section and panel data. Cambridge, Massachusetts Institute of Technology Press.
- YU, Nannan, DE JONG, Martin, STORM, Servaas et MI, Jianing (2013) Spatial spillover effects of transport infrastructure: Evidence from Chinese regions. Journal of Transport Geography, vol. 28, p. 56-66.
- ZUUR, Alain F., HILBE, Joseph M. et IENO, Elena N. (2013) A beginner’s guide to GLM and GLMM with R: A frequentist and Bayesian perspective for ecologists. Newburg, Highland Statistics Limited.