Advertisement

Estimation Results

  • Peter Stalder
Chapter
  • 33 Downloads
Part of the Lecture Notes in Economics and Mathematical Systems book series (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 £.

Keywords

Labor Supply Real Wage Labor Demand Investment Equation Buffer Stock 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 8.
    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. 14.
    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. 19.
    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

Copyright information

© Springer-Verlag Berlin Heidelberg 1991

Authors and Affiliations

  • Peter Stalder
    • 1
  1. 1.Konjunkturforschungsstelle KOF-ETHEidgenössischen Technischen Hochschule ZürichZürichSwitzerland

Personalised recommendations