Least Squares Data Fitting Subject to Decreasing Marginal Returns

Chapter
Part of the Lecture Notes in Electrical Engineering book series (LNEE, volume 229)

Abstract

Let data of a univariate process be given. If the data are related by a sigmoid curve, but the sigmoid property has been lost due to the errors of the measuring process, then the least sum of squares change to the data that provides nonnegative third divided differences is proposed. The method is highly suitable for estimating points on a sigmoid curve of unspecified parametric form subject to increasing marginal returns or subject to diminishing marginal returns. It is a structured quadratic programming calculation, which is solved very efficiently by a special least squares algorithm that takes into account the form of the constraints. Some numerical results illustrate the method on a variety of data sets. Moreover, two applications of the method on real economic data demonstrate its modeling capability. The first one concerns renewable energy consumption data, which exhibit a sigmoid pattern. The second one concerns technological substitutions among the PDP computers to the VAX computers.

Keywords

Approximation Decreasing marginal returns Divided difference Least squares data fitting Optimization  Renewable energy consumption  Sigmoid Technological substitution 

Notes

Acknowledgments

This work was partially supported by the University of Athens under Research Grant 11105.

References

  1. 1.
    Bengisu M, Nekhili R (2006) Forecasting emerging technologies with the aid of science and technology databases. Technological Forecasting and Social Change 73:835–844CrossRefGoogle Scholar
  2. 2.
    Cullinan MP (1990) Data smoothing using non-negative divided differences and l2 approximation. IMA J. of Numerical Analysis 10:583–608MathSciNetMATHCrossRefGoogle Scholar
  3. 3.
    Demetriou IC (2012) "Applications of the discrete least squares 3-convex fit to sigmoid data", Lecture Notes in Engineering and Computer Science: Proceedings of The World Congress on Engineering 2012 WCE 2012, 4–6 July, 2012. U.K, London, pp 285–290Google Scholar
  4. 4.
    Demetriou IC, Lipitakis EA "Least squares data smoothing by nonnegative divided differences", Unpublished manuscriptGoogle Scholar
  5. 5.
    Demetriou IC, Powell MJD (1991) The minimum sum of squares change to univariate data that gives convexity. IMA J. of Numerical Analysis 11:433–448MathSciNetMATHCrossRefGoogle Scholar
  6. 6.
    Dierckx P (1995) Curve and Surface Fitting with Splines. Clarendon Press, OxfordMATHGoogle Scholar
  7. 7.
    Fisher JC, Pry RH (1971) A simple substitution model of technological change. Technological Forecasting and Social Change 2:75–88CrossRefGoogle Scholar
  8. 8.
    Fletcher R (1987) Practical Methods of Optimization. J. Wiley and Sons, Chichester, U.K.MATHGoogle Scholar
  9. 9.
    Fubrycky WJ, Thuesen GJ, Verna D (1998) Economic Decision Analysis, 3rd edn. Prentice Hall, Upper Saddle River, New JerseyGoogle Scholar
  10. 10.
    Gonzalez RC, Wintz P (1987) Digital Image Processing, 2nd edn. Addison Wesley Publishing Company, Reading, MassGoogle Scholar
  11. 11.
    Lindley DV (1985) Making Decisions, 2nd edn. J. Wiley and Sons, London, U.K.Google Scholar
  12. 12.
    McKenna CJ, Rees R (1996) Economics: A Mathematical Introduction. Oxford University Press, New YorkGoogle Scholar
  13. 13.
    Meade N, Islam T (1991) Forecasting with growth curves: an empirical comparison. International Journal of Forecasting 11:199–215CrossRefGoogle Scholar
  14. 14.
    Medawar PB (1941) The Laws of biological growth. Nature 148:772–774CrossRefGoogle Scholar
  15. 15.
    Modis T (1992) Predictions - Society’s Telltale Signature Reveals the Past and Forecasts the Future. Simon and Schuster, New YorkGoogle Scholar
  16. 16.
    Modis T (1993) Technological substitutions in the computer industry. Technological Forecasting and Social Change 43:157–167CrossRefGoogle Scholar
  17. 17.
    Morrison JS (1995) Life-cycle approach to new product forecasting. The Journal of Business Forecasting Methods and Systems 14:3–5Google Scholar
  18. 18.
    Morrison JS (1996) How to use diffusion models in new product forecasting. The Journal of Business Forecasting Methods and Systems 15:6–9Google Scholar
  19. 19.
    Porter ME (1985) Competitive Advantage, Creating and Sustaining Superior Performance. The Free Press, Collier Macmillan Publishers, London, U.KGoogle Scholar
  20. 20.
    Robertson T, Wright FT, Dykstra RL (1988) Order Restricted Statistical Inference. John Wiley and Sons, New YorkMATHGoogle Scholar
  21. 21.
    West JB (1985) Respiratory Physiology - the Essentials, 3rd edn. Williams and Wilkins, BaltimoreGoogle Scholar

Copyright information

© Springer Science+Business Media Dordrecht 2013

Authors and Affiliations

  1. 1.Department of Economics, University of AthensAthensGreece

Personalised recommendations