Double Bootstrap Confidence Intervals in the Two-Stage DEA Approach

A revised version in the Journal of Time Series Analysis, Forthcoming

28 Pages Posted: 7 Feb 2015

See all articles by Dimitris K. Chronopoulos

Dimitris K. Chronopoulos

University of St. Andrews - School of Management

Claudia Girardone

Essex Business School - University of Essex

John C. Nankervis

University of Essex - Department of Accounting, Finance & Management

Date Written: February 5, 2015

Abstract

Contextual factors usually assume an important role in determining firms’ productive efficiencies. Nevertheless, identifying them in a regression framework might be complicated. The problem arises from the efficiencies being correlated with each other when estimated by Data Envelopment Analysis (DEA), rendering standard inference methods invalid. Simar and Wilson (2007) suggest the use of bootstrap algorithms that allow for valid statistical inference in this context. This paper extends their work by proposing a double bootstrap algorithm for obtaining confidence intervals with improved coverage probabilities. It also offers an empirical application using bank-level data. Moreover, acknowledging the computational burden associated with iterated bootstrap procedures, we provide an algorithm based on deterministic stopping rules which is less computationally demanding. Monte Carlo evidence shows considerable improvement in the coverage probabilities after iterating the bootstrap procedure. The results also suggest that percentile confidence intervals perform better than their basic counterpart. The empirical application of the double bootstrap to a sample of European banks suggests that diversified financial conglomerates are more efficient than specialised institutions and the findings are robust to alternative double bootstrap confidence interval methods.

Keywords: Data Envelopment Analysis; Double Bootstrap; Confidence Intervals; Stopping Rules; Two-Stage Approach

JEL Classification: C14, C15, C24, G21

Suggested Citation

Chronopoulos, Dimitris K. and Girardone, Claudia and Nankervis, John C., Double Bootstrap Confidence Intervals in the Two-Stage DEA Approach (February 5, 2015). A revised version in the Journal of Time Series Analysis, Forthcoming, Available at SSRN: https://ssrn.com/abstract=2560866 or http://dx.doi.org/10.2139/ssrn.2560866

Dimitris K. Chronopoulos (Contact Author)

University of St. Andrews - School of Management ( email )

The Gateway
North Haugh
St. Andrews, Fife, Scotland KY16 9SS
United Kingdom

Claudia Girardone

Essex Business School - University of Essex ( email )

Wivenhoe Park
Colchester, CO4 3SQ
United Kingdom
+44 1206874156 (Phone)
+44 1206873429 (Fax)

HOME PAGE: http://www.essex.ac.uk/ebs/staff/profile.aspx?ID=1045

John C. Nankervis

University of Essex - Department of Accounting, Finance & Management ( email )

Wivenhoe Park
Colchester CO4 3SQ
United Kingdom

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
147
Abstract Views
1,335
Rank
358,856
PlumX Metrics