Skip to main content

Part of the book series: Lecture Notes in Economics and Mathematical Systems ((LNE,volume 360))

Abstract

The several versions of the model are estimated with quarterly data over the period 1967.2 to 1985.4 (while the data from 1986.1 to 1989.4 will be used to test post-sample parameter stability). The appropriate likelihood functions, discussed in the preceding section, are concentrated with respect to Ω and maximized using the Davidon-Fletcher-Powell algorithm as implemented in the computer package GQOPT (R.E. Quandt). Starting values for the parameters can be obtained by setting the transformation parameters in [19º] to plausible values, calculating the latent demand and supply variables and regressing them on the explanatory variables of the corresponding equations by single-equation nonlinear least squares. The subsequent ML-estimation accounts for simultaneity of the endogenous variables, cross-equation parameter restrictions and contemporaneous correlation of error terms. Standard errors of the estimated parameters are derived in the usual manner from the approximate inverse of the Hessian at the maximum of ln £.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

eBook
USD 16.99
Price excludes VAT (USA)
  • Available as PDF
  • Read on any device
  • Instant download
  • Own it forever
Softcover Book
USD 16.99
Price excludes VAT (USA)
  • Compact, lightweight edition
  • Dispatched in 3 to 5 business days
  • Free shipping worldwide - see info

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. In the framework of the complete buffer model, which will be estimated over the period 1967.2 to 1985.4 by ML, the demand equation appears with one-quarter lag. To ensure comparability, we estimate it here over the period 1967.1 to 1985.3.

    Google Scholar 

  2. In a linear model one could perform a Lagrange Multiplier test based on the score with respect to the critical elements of R in the constrained model only [see e.g. Engle (1982)]. In a nonlinear model, however, comparing the constrained with the unconstrained model in a Likelihood Ratio test seems saver.

    Google Scholar 

  3. There is an obvious connection to the different concepts of effective trade offers that have been proposed in the fix-price literature: In Clower (1965) or Benassy (1977) agents, when formulating the effective trade offers in a certain market, take into account only the quantity constraints they face in other markets; in Dréze (1975) the agents’ effective trade offers reflect all quantity constraints simulaneously.

    Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Rights and permissions

Reprints and permissions

Copyright information

© 1991 Springer-Verlag Berlin Heidelberg

About this chapter

Cite this chapter

Stalder, P. (1991). Estimation Results. In: Regime Transitions, Spillovers and Buffer Stocks. Lecture Notes in Economics and Mathematical Systems, vol 360. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-46739-4_7

Download citation

  • DOI: https://doi.org/10.1007/978-3-642-46739-4_7

  • Publisher Name: Springer, Berlin, Heidelberg

  • Print ISBN: 978-3-540-54056-4

  • Online ISBN: 978-3-642-46739-4

  • eBook Packages: Springer Book Archive

Publish with us

Policies and ethics