ABSTRACT
Using approximations of the score of the log-likelihood function, we derive moment conditions for estimating spatial regression models, starting with the spatial error model. Our approach results in computationally simple and robust estimators, such as a new moment estimator derived from the first-order approximation obtained by solving a quadratic moment equation, and performs similarly to existing generalized method of moments (GMM) estimators. Our estimator based on the second-order approximation resembles the GMM estimator proposed by Kelejian and Prucha in 1999. Hence, we provide an intuitive interpretation of their estimator. Additionally, we provide a convenient framework for computing the weighting matrix of the optimal GMM estimator. Heteroskedasticity robust versions of our estimators are also proposed. Furthermore, a first-order approximation for the spatial autoregressive model is considered, resulting in a computationally simple method of moment estimator. The performance of the considered estimators is compared in a Monte Carlo study.
ACKNOWLEDGEMENTS
The authors thank the editor-in-chief and two anonymous referees for their very helpful comments and suggestions.
DISCLOSURE STATEMENT
No potential conflict of interest was reported by the authors.
Notes
1. For some recent applications, see Lin (Citation2010), Piras, Postiglione, and Aroca (Citation2012), de Dominicis, Florax, and de Groot (Citation2013), Kelejian, Murrell, and Shepotylo (Citation2013), Miguelez and Moreno (Citation2013), and Brady (Citation2014). For a list of applications from 1991 until 2007, see Kelejian and Prucha (Citation2010b).
2. Important contributions to the development of the GMM framework include Kelejian and Prucha (Citation2001, Citation2004, Citation2007, Citation2010a, Citation2010b), Lee (Citation2002, Citation2003, Citation2004, Citation2007a, Citation2007b), Kelejian, Prucha, and Yuzefovich (Citation2004), Kapoor et al. (Citation2007), Fingleton (Citation2008a, Citation2008b), Fingleton and Le Gallo (Citation2008a, Citation2008b), Arnold and Wied (Citation2010), Arraiz, Drukker, Kelejian, and Prucha (Citation2010), Elhorst (Citation2010b), Lee and Liu (Citation2010), Lin and Lee (Citation2010), Liu and Lee (Citation2010, Citation2013), Liu et al. (Citation2010), Baltagi and Liu (Citation2011), Drukker et al. (Citation2013), Wang and Lee (Citation2013), Kelejian and Piras (Citation2014), Lee and Yu (Citation2014, Citation2016), Qu and Lee (Citation2015), and Qu, Wang, and Lee (Citation2016).
3. Elhorst (Citation2010a) provides a comprehensive discussion and comparison of different types of spatial regression models, whereas spatial panel models are analyzed by Elhorst (Citation2003), Kapoor et al. (Citation2007) and Elhorst (Citation2012).
5. In the fourth section, we will split that linear combination into two moment conditions, yielding an over-identified GMM version of the MLAM2 estimator.
7. For convenience, we assume to be known. As usual, a consistent estimator can be obtained from the residuals of the model.
8. We are grateful to an anonymous referee for pointing out this potential problem with GMM estimates in the SDM.
11. These estimators differ only in the second element of the moment vector given in (27). In the case of a symmetrical spatial weight matrix, they are equal, since then .
Lin, X. (2010). Identifying peer effects in student academic achievement by spatial autoregressive models with group unobservables. Journal of Labor Economics, 28, 825–860. doi: 10.1086/653506 Piras, G., Postiglione, P., & Aroca, P. (2012). Specialization, R&D and productivity growth: Evidence from EU regions. Annals of Regional Science, 49, 35–51. doi: 10.1007/s00168-010-0424-2 de Dominicis, L., Florax, R. J. G. M., & de Groot, H. L. F. (2013). Regional clusters of innovative activity in Europe: Are social capital and geographical proximity key determinants? Applied Economics, 45, 2325–2335. doi: 10.1080/00036846.2012.663474 Kelejian, H. H., Murrell, P., & Shepotylo, O. (2013). Spatial spillovers in the development of institutions. Journal of Development Economics, 101, 297–315. doi: 10.1016/j.jdeveco.2012.12.003 Miguelez, E., & Moreno, R. (2013). Research networks and inventors’ mobility as drivers of innovation: Evidence from Europe. Regional Studies, 47, 1668–1685. doi: 10.1080/00343404.2011.618803 Brady, R. R. (2014). The spatial diffusion of regional housing prices across U.S. States. Regional Science and Urban Economics, 46, 150–166. doi: 10.1016/j.regsciurbeco.2014.04.003 Kelejian, H. H., & Prucha, I. R. (2010b). Specification and estimation of spatial autoregressive models with autoregressive and heteroskedastic disturbances. Journal of Econometrics, 157, 53–67. doi: 10.1016/j.jeconom.2009.10.025 Kelejian, H. H., & Prucha, I. R. (2001). On the asymptotic distribution of the Moran I test statistic with applications. Journal of Econometrics, 104, 219–257. doi: 10.1016/S0304-4076(01)00064-1 Kelejian, H. H., & Prucha, I. R. (2004). Estimation of simultaneous systems of spatially interrelated cross sectional equations. Journal of Econometrics, 118, 27–50. doi: 10.1016/S0304-4076(03)00133-7 Kelejian, H. H., & Prucha, I. R. (2007). HAC estimation in a spatial framework. Journal of Econometrics, 140, 131–154. doi: 10.1016/j.jeconom.2006.09.005 Kelejian, H. H., & Prucha, I. R. (2010a). Spatial models with spatially lagged dependent variables and incomplete data. Journal of Geographical Systems, 12, 241–257. doi: 10.1007/s10109-010-0109-5 Kelejian, H. H., & Prucha, I. R. (2010b). Specification and estimation of spatial autoregressive models with autoregressive and heteroskedastic disturbances. Journal of Econometrics, 157, 53–67. doi: 10.1016/j.jeconom.2009.10.025 Lee, L. (2002). Consistency and efficiency of least squares estimation for mixed regressive, spatial autoregressive models. Econometric Theory, 18, 252–277. doi: 10.1017/S0266466602182028 Lee, L.-F. (2003). Best spatial two-stage least squares estimators for a spatial autoregressive model with autoregressive disturbances. Econometric Reviews, 22, 307–335. doi: 10.1081/ETC-120025891 Lee, L.-F. (2004). Asymptotic distributions of quasi-maximum likelihood estimators for spatial autoregressive models. Econometrica, 72, 1899–1925. doi: 10.1111/j.1468-0262.2004.00558.x Lee, L.-F. (2007a). GMM and 2SLS estimation of mixed regressive, spatial autoregressive models. Journal of Econometrics, 137, 489–514. doi: 10.1016/j.jeconom.2005.10.004 Lee, L. (2007b). The method of elimination and substitution in the GMM estimation of mixed regressive, spatial autoregressive models. Journal of Econometrics, 140, 155–189. doi: 10.1016/j.jeconom.2006.09.006 Kelejian, H. H., Prucha, I. R., & Yuzefovich, Y. (2004). Instrumental variable estimation of a spatial autoregressive model with autoregressive disturbances: Large and small sample results. In Spatial and spatiotemporal econometrics, advances in econometrics (Vol. 18, pp. 163–198). New York: Elsevier. Kapoor, M., Kelejian, H. H., & Prucha, I. R. (2007). Panel data models with spatially correlated error components. Journal of Econometrics, 140, 97–130. doi: 10.1016/j.jeconom.2006.09.004 Fingleton, B. (2008a). A generalized method of moments estimator for a spatial model with moving average errors, with application to real estate prices. Empirical Economics, 34, 35–57. doi: 10.1007/s00181-007-0151-4 Fingleton, B. (2008b). A generalized method of moments estimator for a spatial panel model with an endogenous spatial Lag and spatial moving average errors. Spatial Economic Analysis, 3, 27–44. doi: 10.1080/17421770701774922 Fingleton, B., & Le Gallo, J. (2008a). Estimating spatial models with endogenous variables, a spatial lag and spatially dependent disturbances: Finite sample properties. Papers in Regional Science, 87, 319–339. doi: 10.1111/j.1435-5957.2008.00187.x Fingleton, B., & Le Gallo, J. (2008b). Finite sample properties of estimators of spatial models with autoregressive, or moving average, disturbances and system feedback. Annales d’Economie et de Statistique, 87/88, 39–62. Arnold, M., & Wied, D. (2010). Improved GMM estimation of the spatial autoregressive error model. Economics Letters, 108, 65–68. doi: 10.1016/j.econlet.2010.04.012 Arraiz, I., Drukker, D. M., Kelejian, H. H., & Prucha, I. R. (2010). A spatial Cliff–Ord-type model with heteroskedastic innovations: Small and large sample results. Journal of Regional Science, 50, 592–614. doi: 10.1111/j.1467-9787.2009.00618.x Elhorst, J. P. (2010b). Dynamic panels with endogenous interaction effects when T is small. Regional Science and Urban Economics, 40, 272–282. doi: 10.1016/j.regsciurbeco.2010.03.003 Lee, L.-F., & Liu, X. (2010). Efficient GMM estimation of high order spatial autoregressive models with autoregressive disturbances. Econometric Theory, 26, 187–230. doi: 10.1017/S0266466609090653 Lin, X., & Lee, L.-F. (2010). GMM estimation of spatial autoregressive models with unknown heteroskedasticity. Journal of Econometrics, 157, 34–52. doi: 10.1016/j.jeconom.2009.10.035 Liu, X., & Lee, L.-F. (2010). GMM estimation of social interaction models with centrality. Journal of Econometrics, 159, 99–115. doi: 10.1016/j.jeconom.2010.04.009 Liu, X., & Lee, L.-F. (2013). Two-stage least squares estimation of spatial autoregressive models with endogenous regressors and many instruments. Econometric Reviews, 32, 734–753. doi: 10.1080/07474938.2013.741018 Liu, X., Lee, L.-F., & Bollinger, C. R. (2010). An efficient GMM estimator of spatial autoregressive models. Journal of Econometrics, 159, 303–319. doi: 10.1016/j.jeconom.2010.08.001 Baltagi, B. H., & Liu, L. (2011). Instrumental variable estimation of a spatial autoregressive panel model with random effects. Economics Letters, 111, 135–137. doi: 10.1016/j.econlet.2011.01.016 Drukker, D. M., Egger, P., & Prucha, I. R. (2013). On two-step estimation of a spatial autoregressive model with autoregressive disturbances and endogenous regressors. Econometric Reviews, 32, 686–733. doi: 10.1080/07474938.2013.741020 Wang, W., & Lee, L.-F. (2013). Estimation of spatial panel data models with randomly missing data in the dependent variable. Regional Science and Urban Economics, 43, 521–538. doi: 10.1016/j.regsciurbeco.2013.02.001 Kelejian, H. H., & Piras, G. (2014). Estimation of spatial models with endogenous weighting matrices, and an application to a demand model for cigarettes. Regional Science and Urban Economics, 46, 140–149. doi: 10.1016/j.regsciurbeco.2014.03.001 Lee, L.-F., & Yu, J. (2014). Efficient GMM estimation of spatial dynamic panel data models with fixed effects. Journal of Econometrics, 180, 174–197. doi: 10.1016/j.jeconom.2014.03.003 Lee, L.-F., & Yu, J. (2016). Identification of spatial Durbin panel models. Journal of Applied Econometrics, 31, 133–162. doi: 10.1002/jae.2450 Qu, X., & Lee, L.-F. (2015). Estimating a spatial autoregressive model with an endogenous spatial weight matrix. Journal of Econometrics, 184, 209–232. doi: 10.1016/j.jeconom.2014.08.008 Qu, X., Wang, X., & Lee, L.-F. (2016). Instrumental variable estimation of a spatial dynamic panel model with endogenous spatial weights when T is small. Econometrics Journal, 19, 261–290. doi: 10.1111/ectj.12069 Elhorst, J. P. (2010a). Applied spatial econometrics: Raising the bar. Spatial Economic Analysis, 5, 9–28. doi: 10.1080/17421770903541772 Elhorst, J. P. (2003). Specification and estimation of spatial panel data models. International Regional Science Review, 26, 244–268. doi: 10.1177/0160017603253791 Kapoor, M., Kelejian, H. H., & Prucha, I. R. (2007). Panel data models with spatially correlated error components. Journal of Econometrics, 140, 97–130. doi: 10.1016/j.jeconom.2006.09.004 Elhorst, J. P. (2012). Dynamic spatial panels: Models, methods, and inferences. Journal of Geographical Systems, 14, 5–28. doi: 10.1007/s10109-011-0158-4 Additional information
Funding
Christoph Wigger gratefully acknowledges the financial support provided by the German Research Foundation in the context of the Priority Program 1764 ‘The German Labor Market in a Globalized World - Challenges through Trade, Technology, and Demographics’.