Abstract
The applied statistician/econometrician who has to select an estimation method needs information about the relative performance of two or more estimators under empirically relevant conditions. The term “empirically relevant conditions” refers e.g. to small samples, to the presence of specification errors and to an econometric model which ist “complicated” in the following sense: It contains a large number of interdependent variables (equations), it is dynamic; its exogenous variables are not free of measurement errors, are autocorrelated and the data are collinear; at least some observable variables are cointegrated; the disturbances in the equations are autocorrelated within and across equations and are perhaps also heteroskedastic. Even more complicated are models which also contain present and future model consistent (= “rational”) expectations variables.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Aigner, D.J.: A brief introduction to the methodology of optimal experimental design, in: Journal of Econometrics 11, 1979, pp. 7–26.
Chase MacRae, E.: Optimal experimental design for dynamic econometric models, in: Annals of Economic and Social Measurement 6, 1977, pp. 399–405.
Conlisk, J.: Optimal response surface design in Monte Carlo sampling experiments, in: Annals of Economic and Social Measurement 3/3, 1974, pp. 463–473.
Conlisk, J., and Watts, H.: A model for optimizating experimental design for estimating response surfaces, in: Journal of Econometrics 11, 1979, pp. 27–42.
Freimann, K.-D.: Parameterschätzung in interdependenten dynamischen ökonometrischen Gleichungssystemen mit autokorrelierten Störvariablen - eine Monte-Carlo-Studie, Krefeld 1984.
Freimann, K.-D., and Gruber, J.: Small sample properties of three estimators in dynamic simultaneous equations models with disturbances generated by a vector autoregressive process - a Monte Carlo study, in: Milo, W., and Miszcyiński, M. (eds.): Macromodels and forecasts of national economy, Polish Scientific Publishers, Warsow, Lodz, 1986, pp. 49–80.
Gruber, J.: Neuantrag auf eine Sachbeihilfe im Rahmen des Schwerpunktprogramms „Monetäre MakroÖkonomik“der DFG: Teil 1, Simulationsgrundlagen für ökonometrische Untersuchungen, FernUniversität Hagen, 13. 5. 1986.
Gruber, J., and Freimann, K.-D.: Combined response surface regressions in Monte Carlo studies of small sample properties of estimators: theory and an application, Discussion Paper No. 108, Department of Economics, University of Hagen, Hagen 1986.
Gruber, J., Fusek, I., and Fusková, L.: Evaluating the relative performance of estimators by means of combined response surfaces: the case of EV estimators and OLS, in: Jahrbücher für Nationalökonomie und Statistik 203, 1987, pp. 507–516.
Hendry, D.F.: The behaviour of inconsistent instrumental variables estimators in dynamic systems with autocorrelated errors, in: Journal of Econometrics 7, 1979, pp. 295–314.
Hendry, D.F.: Monte Carlo experimentation in econometrics, in: Griliches, Z., and Intriligator, M.D. (eds.): Handbook of Econometrics, Volume II, Amsterdam and Oxford, 1984, pp. 937–976.
Hendry, D.F. (with D. Qin and C. Favero): Lectures on econometric methodology, Nuffield College, Oxford (175 pp.; revised version to be published by Oxford University Press, 1991 ).
Hendry, D.F., and Harrison, R.W.: Monte Carlo methodology and the small sample behaviour of ordinary and two-stage least squares, in: Journal of Econometrics 2, 1974, pp. 151–174.
Hendry, D.F., and Srba, F.: The properties of autoregressive instrumental variables estimators in dynamic systems, in: Econometrica 45, 1977, pp. 969–990.
Hillier, G.H., Kinal, T.W., and Srivastava, V.K.: On the moments of ordinary least squares and instrumental variables estimators in a general structural equation, in: Econometrica 52, 1984, pp. 185–202.
Kendall, M.G.: Rank correlation methods (4th ed.), London 1975.
Kleijnen, J.P.C.: Statistical techniques in simulation, part II, New York 1975.
Krämer, W.: Eine Rehabilitation der gewöhnlichen Kleinst-Quadrate-Methöde als Schätzverfahren in der Ökonometrie, Frankfurt 1980.
Leontief, W.: Theoretical assumptions and nonobserved facts, in: The American Economic Review 61, 1971, pp. 1–7.
Mariano, R.S.: Analytical small-sample distribution theory in econometrics: the simultaneous equations case, in: International Economic Review 23, 1982, pp. 503–533.
Mittag, H.-J.: Modifizierte Kleinst-Quadrat-Schätzung im Modell mit fehlerbehafteten Daten, Mathematical Systems in Economics, Band 109, Frankfurt 1987.
Mittag, H.-J.: Estimating parameters in a simple errors-in-variables model: a new approach based on finite sample distribution theory, in: Statistical Papers 30, 1989, pp. 133–140.
Phillips, P.C.B.: Exact small sample theory in the simultaneous equations model, in: Griliches, Z., and Intriligator, M.D. (eds.): Handbook of Econometrics, Volume I, Amsterdam 1983, pp. 449–516.
Phillips, P.C.B.: The exact distribution of LIML: I, in: International Economic Review 25,1984, pp. 249–261.
Phillips, P.C.B.: The exact distribution of LIML: II, in: International Economic Review 26, 1985, pp. 21–36.
Rinne, H.: A method of choosing additional sets of observations in multiple linear regression models to overcome multieollinearity, in: Gruber, J. (ed.): Multicollinearity and biased estimation, Applied Statistics and Econometrics 27, Göttingen 1984, pp. 25–44.
Schaak, J.: Abgrenzung des zulässigen Parameterraumes in statistischen Experimenten auf dem Computer, Seminararbeit in Statistik, FernUniversität Hagen, Mai 1990.
Shelton, J.T., Khuri, A.I., and Korwell, J.A.: Selecting check points for testing lack of fit in response surface models, in: Technometrics 25, 1983, pp. 357–365.
Steinberg, D.M., and Hunter, W.G.: Experimental design: review and comment, in: Technometrics 26, 1984, pp. 71–97 and 128–130.
Trampenau, P.: Kleinstichprobeneigenschaften ausgewählter Schätzverfahren im multivariaten Fehler-in-den-Variablen-Modell, Diskussionsbeitrag 135, Fachbereich Wirtschaftswissenschaft, FernUniversität Hagen, Hagen 1988.
Wawrzynek, J.: On asymptotically optimal experimental design for multidimensional linear regression, in: Przeglad Statystyczny XXVIII, 1981, pp. 271–282.
Wawrzynek, J.: Zur Versuchsplanung für die lineare Regression, Paper, University of Economics, Wroclaw, on which a lecture at the University of Hagen (FRG) on Dec. 14,1983, was based.
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1991 Springer-Verlag Berlin Heidelberg
About this chapter
Cite this chapter
Gruber, J. (1991). Improving the Generality of Results from Monte Carlo Studies of Small Sample Properties of Estimators. In: Fandel, G., Gehring, H. (eds) Operations Research. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-76537-7_11
Download citation
DOI: https://doi.org/10.1007/978-3-642-76537-7_11
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-76539-1
Online ISBN: 978-3-642-76537-7
eBook Packages: Springer Book Archive