Summary
The ultimate objective of this paper is to arrive at an operational testing procedure enabling us to diagnose the causal or non-causal nature of an econometric relationship in a linear framework. The profferred approach to the problem relies on a close examination of causal issues in simultaneous equation models and on a sensible definition of the bidirectionality concept inherently associated with feedback mechanisms.
After recalling the essentials of causal structure's algebra, the article:
-
i)
provides a closed-form expression, in terms of a Hadamard product of forward and backward (i.e., feedback) effects, for two-way connections among the variables of a linear equation;
-
ii)
sets the issue of verifying a causality conjecture regarding a model's relationship in the context of significance test theory;
-
iii)
builds up, in a constrained maximum-likelihood setting, an effective causality test of the Lagrange-multipler (scoring) class.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Aigner D. J. andA. Zellner, eds. (1988), Causality,A supplement to the Journal of Econometrics, Annals 1988/3.
Ancot J. P., ed. (1984),Analysing the Structure of Econometric Models, The Hague, Nijhoff.
Bentzel R. andB. Hansen (1954), On recursiveness and interdependency in economic models,Review of Economic Studies, XXII, 153–168.
Breusch T. S. andA. R. Pagan (1980), The Lagrange multipler test and its applications to model specifications in econometrics,Review of Economic Studies, XLVII, 239–253.
Cappuccio N. andR. Orsi (1991),Econometria, Il Mulino, Bologna.
Carlucci F., ed. (1985),Analisi della causalità e modelli econometrici, 2 voll., ISCO, Roma.
Dagum C. (1969), Structural performance,Zeitschrift fur die Gesamte Staatswissenschaft, 211–235.
D'Azzo J. andC. H. Houpis (1966),Feedback Control System Analysis and Synthesis, Mc-Graw Hill, New York.
Engle F. (1982), A general approach to Lagrange multipler model diagnosis,Journal of Econometrics, 20, 83–104.
Faliva M. (1987),Econometria: Principi e Metodi, UTET, Torino.
Faliva M. (1991), L'analisi dei modelli econometrici nell'ambito della teoria dei sistemi, in M. Faliva ed.,Il ruolo dell'econometria nell'ambito delle scienze economiche, 115–126, Il Mulino, Bologna.
Faliva M. (1992), Recursiveness vs. interdependence in linear econometric models: a comprehensive analysis for the linear case,Journal of the Italian Statistical Society, I, 335–357.
Faliva M. (1994), Causality and interdependence in linear econometric models,forthcoming.
Godfrey L. G. (1988),Misspecification Tests in Econometrics, Cambridge University Press, New York.
Gilli M. (1992), Causal ordering and beyond,International Economic Review, 33, 957–965.
Granger C. W. J. (1980), Testing for causality: a personal viewpoint,Journal of Economic Dynamics and Control, 2, 329–352.
Hausman J. A. (1978), Specification tests in econometrics,Econometrica, 46, 1251–1271.
Joreskog K. G. andH. Wold, eds. (1982),Systems under Indirect Observation, 2 voll., North-Holland, Amsterdam.
Mosbaek E. andH. Wold (1970),Interdependent Systems: Structure and Estimation, North-Holland, Amsterdam.
Pasinetti L. (1965), Causalità ed interdipendenza nell'analisi econometrica e nella teoria economica,Annali Università Cattolica A.A., 1964–65, 233–250, Vita e Pensiero, Milano.
Rossier E. (1980),Economie structurale, Economica, Parigi.
Smith R. J. (1985), Wald tests for the independence of stochastic variables and disturbance of a single linear stochastic simultaneous equation,Economic letters, 17, 87–90.
Strotz H. andH. Wold (1960), Recursive vs. non recursive systems: an attempt at synthesis;Econometrica, 28, 417–427.
Wold H. (1964a),Econometric Model Building, North-Holland, Amsterdam.
Wold H. (1964b),La technique des modèles, Science Humaines, Monaco.
Wu De-Min (1973), Alternative tests of independence between stochastic regresors and disturbances,Econometrica, 41, 733–750.
Zellner A. (1979), Causality and econometrics, inK. Brunner andMeltzer A. H., eds.,Three Aspects of Policy and Policy Making, 9–54, North-Holland, Amsterdam.
Zoia M. G. (1994), ANRINT/1: Software per l'analisi dei meccanismi ricorsivi ed interdipendenti dei modelli econometrici,Contributi di Ricerca in Econometria, 94.02, Università Cattolica, Milano.
Author information
Authors and Affiliations
Additional information
Invited paper at the Conference held in Bologna, Italy, 27–28 May 1993, on «Statistical Tests: Methodology and Econometric Applications».
Support from the Italian Research Council (CNR) is gratefully acknowledged. The paper is the result of a joint research of M. Faliva and M. G. Zoia. The two authors share the responsability for Section 3; Sections 1 and 2 are due to M. Faliva and Section 4 is due to M. G. Zoia.
Rights and permissions
About this article
Cite this article
Faliva, M., Zoia, M.G. Detecting and testing causality in linear econometric models. J. It. Statist. Soc. 3, 61–76 (1994). https://doi.org/10.1007/BF02589041
Issue Date:
DOI: https://doi.org/10.1007/BF02589041