A new test for the Box-Cox transformation model: An analysis of length of hospital stay for diabetes patients in Japan

Kazumitsu Nawata () and Koichi Kawabuchi ()
Additional contact information
Kazumitsu Nawata: University of Tokyo
Koichi Kawabuchi: Tokyo Medical and Dental University

Economics Bulletin, 2014, vol. 34, issue 1, 324-332

Abstract: The Box-Cox (1964) transformation model (BC model) is widely used to examine various problems. The likelihood function under the normality assumption is misspecified, and the maximum likelihood estimator (BC MLE) cannot in general be consistent. However, under the “small sigma” assumption described in Bickel and Doksum (1981), the BC MLE can be consistent. It is, therefore, necessary to test whether this assumption is satisfied when the BC model is used. In this paper, we propose a new test of whether or not the BC MLE can be used based on the estimator proposed by Nawata (2013). We then analyze length of hospital stay for type 2 diabetes patients hospitalized for educational programs about managing diabetes at home. A dataset of 970 patients collected from 27 general hospitals in Japan is used.

Keywords: Box-Cox transformation model; power transformation model; diabetes; length of stay (LOS) (search for similar items in EconPapers)
JEL-codes: C2 I0 (search for similar items in EconPapers)
Date: 2014-02-28
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (1)

Downloads: (external link)
http://www.accessecon.com/Pubs/EB/2014/Volume34/EB-14-V34-I1-P32.pdf (application/pdf)

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:ebl:ecbull:eb-13-00735

Access Statistics for this article

More articles in Economics Bulletin from AccessEcon
Bibliographic data for series maintained by John P. Conley ().