Regression analyses of a spatially distributed variable is a common technique in urban and regional planning. Frequently, however, the disturbance terms are not completely independent but exhibit spatial autocorrelation. A method is proposed for taking account of this spatial autocorrelation and is illustrated by means of an application to a Belgian labour-demand function.
Get full access to this article
View all access options for this article.
References
1.
AndersonT. W., 1948, “On the theory of testing serial correlation”, Skandinavisk Aktuarietidskrift, 31, 88–116.
2.
CliffA. D.OrdJ. K., 1973, Spatial Autocorrelation (Pion, London).
3.
DurbinJ., 1960, “Estimation of parameters in time series regression models”, Journal of the Royal Statistical Society, Series B, 22, 139–153.
4.
Institut National de Statistiques (INS), “Rapports annuels”, Bruxelles.
5.
MeadR., 1967, “A mathematical model for the estimation of interplant competition”, Biometrics, 23, 189–205.
6.
Office National de Sécurité Sociale (ONSS), “Rapports annuels”, Bruxelles.
7.
TermoteM., 1972 (avec la collaboration de Bodson, P., Lievens, J.), “Une fonction de demande régionale de travail—Belgique 1964–1969”, Département Travail, U.C.L., 1972, pp. 88.
8.
TheilH., 1965, “The analysis of disturbances in regression residuals”, Journal of the American Statistical Association, 60, 1067–1079.
9.
TheilH., 1971, Principles of Econometrics (John Wiley, New York).
10.
Van GrevelingheG., 1973, Économétrie, Collection Méthodes (Hermann, Paris).
11.
WhittleP., 1954, “On stationary processes in the plane”, Biometrika, 41, 434–449.
