The Restricted Least Squares Stein-Rule in gretl

Lee Adkins

No 1305, Economics Working Paper Series from Oklahoma State University, Department of Economics and Legal Studies in Business

Abstract: The paper documents a Gretl function package that is used for the Restricted Least Squares (RLS) Stein-rule estimator. Judge and Bock (1981, pp. 240-42) proposed a family of Stein-rule estimators that dominates the MLE of in the CNLRM under weighted quadratic loss. The estimator is a linear combination of the unrestricted and restricted MLEs, where the degree of shrinkage is controlled using a conventional Wald test of the implied hypothesis restrictions. The Gretl function computes the positive-part version of the RLS Stein-rule that allows users to specify the desired linear restrictions on the model and to select a loss function under which to compute the RLS-Stein rule. In the absence of specifi c prior information about parameter values Lindley's version of the James-Stein rule is particularly attractive; accordingly, it is off ered as a user speci fied option. The fi nal version of the paper will also provide options for computing bootstrap standard errors [see Adkins (1990); Adkins and Hill 1990a)]. A simple Monte Carlo simulation is performed to explore the risk characteristics of the RLS-Stein rule vs. those of pretest, restricted mle, and unrestricted mle. All of the computations are preformed in gretl.

Keywords: Shrinkage estimation; prior information; gretl (search for similar items in EconPapers)
JEL-codes: C01 (search for similar items in EconPapers)
Pages: 15 pages
Date: 2013-06
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://business.okstate.edu/site-files/docs/ecls-working-papers/OKSWPS1305.pdf (application/pdf)
Our link check indicates that this URL is bad, the error code is: 404 Not Found

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:okl:wpaper:1305

Access Statistics for this paper

More papers in Economics Working Paper Series from Oklahoma State University, Department of Economics and Legal Studies in Business Contact information at EDIRC.
Bibliographic data for series maintained by Harounan Kazianga ().