Abstract
Nonparametric Kernel methods have been discussed by Vinod and Ullah (1988), who develop an estimator of the partial derivatives of conditional expectations. These have been applied in Econometrics for estimation of elasticities and other constructs related to partial derivatives. Applications of those ideas need not be restricted to the first pariials. This paper is concerned with estimation of second partials with similar kernel methods. We attempt to empirically study Marshall’s ‘law’ of the elasticity of demand which claims that elasticity increases with price.
John McMillan and Aman Ullah thank the National Science Foundation and SSHRCC, respectively, for research support. This paper is an extension of a paper presented at the Third Canadian Econometric Study Group Conference in Montreal, Canada, September 1986. The authors thank Robin Carter and Preston McAfee for comments.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Notes
Marshall (1920), p. 87. An account of Marshall’s invention of the concept of elasticity is given by Keynes (1963), who described it as one of Marshall’s most important contributions to economics.
In an earlier paper Gasser and Müller (1984) considered the second derivative in the case of regression with fixed design variables. In the related work Rilstone (1987) and Rilstone and Ullah (1987) have considered the derivative estimation by using the finite difference method.
We ignore here possible simultaneity problems arising from the interaction of supply and demand. This can be given both an economic and a statistical justification. First, the existence of a production lag may mean that supply does not depend on current price; instead, it depends on a prediction about price made some time in the past. Second, Kramer (1984) has shown that there is no asymptotic simultaneous equations bias when the regressors are trended.
We note that the estimates of second order partial may be biased due to the normal kernel considered in this paper. This bias could however be eliminated by using kernels which are not necessarily positive; see Bartlett (1963). One such kernel is g 3(w) = 2-1 K(w) (3-w 2), another kernel among many in Vinod (1987) is g 7(w) = (1 /48)(105-105w 2+21w 4-w 6)K(w), where K(w) is the same as in (5). It can be shown that g 3 satisfies μ′1 = μk = 0 for its first three moments; and similarly g 7 satisfies μ′1 = μk = 0 for k = 2,3,4,5,6,7. These will be the subject of future study.
References
Bartlett, M. S.: 1963, ’statistical Estimation of Density Function’, Shankhya, Ser. A., 25, 245–254.
Gasser, T., and H. G. Müllen 1984, ‘Estimating Regression Functions and their Derivatives by the Kernel Method’, Scandinavian Journal of Statistics, 11, 171–185.
Keynes, J. M.: 1963, Essays in Biography, Norton, New York, New York, U.S.A.
Krämer, W.: 1984, ‘On the Consequences of Trend for Simultaneous Equation Estimation’, Economics Letters, 14, 23–30.
Marshall, A.: 1920, Principles of Economics, 8th Ed., Macmillan, London, England.
Novshek, W.: 1985, ‘On the Existence of Cournot Equilibrium’, Review of Economic Studies, 52, January, 85–98.
Prakasa Rao, B. L. S.: 1983, Nonparametric Functional Estimation, Academic Press, New York, New York, U.S.A.
Rilstone, P., and A. Ullah: 1987, ‘Nonparametric Estimation of Response Coefficients’, mimeo, University of Western Ontario, Canada.
Rilstone, P.: 1987, ‘Nonparametric Partial Devrivative Estimation’, Ph.D. Thesis, University of Western Ontario, Canada.
Rosenblatt, M.: 1956, ‘Remarks on some nonparametric estimates of density’, Annals of Mathematical Statistics, 27, 832–837.
Singh, R., A. Ullah, and R. A. L. Carter: 1987, ‘Nonparametric Inference in Econometrics: New Applications’, in I. MacNeil and G. Umphrey (eds.), Time Series and Econometric Models, D. Reidell Publishing, Boston, Massachusetts, U.S.A.
Theil, H.: 1976, Theory and Measurement of Consumer Demand, Vol. 2, North Holland, New York, New York, U.S.A.
Vinod, H. D.: 1987, ’simple New Densities with Zero Low Order Moments’, Discussion Paper ES188, Department of Econometrics, University of Manchester, Manchester, U.K.
Vinod, H. D., and A. Ullah: 1988, ‘Flexible Production Estimation of Nonparametric Kernel Estimators’, in T. B. Fomby and G. F. Rhodes (eds.), Advances in Econometrics: Robust and Nonparametric Statistical Inference, JAI Press, forthcoming.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1989 Springer Science+Business Media Dordrecht
About this chapter
Cite this chapter
McMillan, J., Ullah, A., Vinod, H.D. (1989). Estimation of the Shape of the Demand Curve by Nonparametric Kernel Methods. In: Raj, B. (eds) Advances in Econometrics and Modelling. Advanced Studies in Theoretical and Applied Econometrics, vol 15. Springer, Dordrecht. https://doi.org/10.1007/978-94-015-7819-6_6
Download citation
DOI: https://doi.org/10.1007/978-94-015-7819-6_6
Publisher Name: Springer, Dordrecht
Print ISBN: 978-90-481-4048-0
Online ISBN: 978-94-015-7819-6
eBook Packages: Springer Book Archive