Discussion Paper
No. 2007-13 | March 26, 2007
Christian B. Hansen, James B. McDonald and Panayiotis Theodossiou
Some Flexible Parametric Models for Partially Adaptive Estimators of Econometric Models

Abstract

This paper discusses three families of flexible parametric probability density functions: the skewed generalized t, the exponential generalized beta of the second kind, and the inverse hyperbolic sin distributions. These families allow quite flexible modeling the first four moments of a distribution and could be considered in modeling a wide variety of economic problems. We illustrate their use in a simple regression model with a simulation study that demonstrates that the use of the flexible distributions may result in significant efficiency gains relative to more conventional regression procedures, such as ordinary least squares or least absolute deviations regression, without a suffering from a large efficiency loss when errors are Gaussian.

JEL Classification:

C13, C14, C15

Links

Cite As

[Please cite the corresponding journal article] Christian B. Hansen, James B. McDonald, and Panayiotis Theodossiou (2007). Some Flexible Parametric Models for Partially Adaptive Estimators of Econometric Models. Economics Discussion Papers, No 2007-13, Kiel Institute for the World Economy. http://www.economics-ejournal.org/economics/discussionpapers/2007-13


Comments and Questions



anonymous - Referee Report
April 23, 2007 - 16:17
see attached file
[ Download attached file ]

anonymous - Referee Report
April 25, 2007 - 10:13
The paper is relevant to current concerns and issues in econometrics, especially in finance where data and regression errors have a tendency to have thick-tailed and skewed distributions. The topic of the paper is verytimely and well-written, but there is little in the paper by way of examples in economics and finance that frame the value of the contribution of the paper. The paper includes a simulation that demonstrates the improvement inestimation efficiency from specifying the flexible distributions that they review for the likelihood functions of the regressions. Since the paper seems to be structured as a note, I would not expect them to do an empiricalanalysis with a dataset, but it would be helpful to cite and discuss an application or two that is already in the literature.