Approximations and inference for envelopment estimators of production frontiers

Cinzia Daraio () and Leopold Simar
Additional contact information
Cinzia Daraio: Sapienza University of Rome

Journal of Productivity Analysis, 2024, vol. 62, issue 2, No 4, 197-215

Abstract: Abstract Nonparametric methods have been commonly used to assess the performance of both private and public organizations. Among them, the most popular ones are envelopment estimators such as Free Disposal Hull (FDH) or Data Envelopment Analysis (DEA), which estimate the attainable sets and their efficient boundaries by enveloping the cloud of observed units in the appropriate input-output space. However, these nonparametric envelopment techniques do not provide estimates of marginal products and other coefficients of economic interest. This paper presents a new approach that provides local estimates of all the desired partial derivatives and economic coefficients, which complement and complete the analysis based on nonparametric envelopment estimators. We improve nonparametric estimators by estimating nonparametrically smoothed efficient boundaries and providing derivatives and other coefficients without having to assume any parametric structure for the frontier and the inefficiency distribution. Our approach offers several advantages, such as a flexible nonparametric adjustment of the efficient frontier based on local linear models; a general multivariate efficiency model based on directional distances where one can choose the desired benchmark direction; the possibility of assessing the impact of external-environmental variables; a bootstrap-based statistical inference for deriving confidence intervals on the estimated coefficients for nonparametric and robust frontier approximations; the possibility of including factors aggregating inputs or outputs and recovering the estimated coefficients in the original units. To demonstrate the usefulness of the proposed approach, we provide an illustration in the field of education, where economic coefficients are important but the parametric assumptions have been questioned.

Keywords: Data envelopment analysis; Partial frontiers; Directional distances; Linear approximations; Local linear approximations (search for similar items in EconPapers)
JEL-codes: C1 C13 C14 (search for similar items in EconPapers)
Date: 2024
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
http://link.springer.com/10.1007/s11123-024-00726-2 Abstract (text/html)
Access to full text is restricted to subscribers.

Related works:
Working Paper: Approximations and inference for envelopment estimators of production frontiers (2024)
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:kap:jproda:v:62:y:2024:i:2:d:10.1007_s11123-024-00726-2

Ordering information: This journal article can be ordered from
http://www.springer. ... cs/journal/11123/PS2

DOI: 10.1007/s11123-024-00726-2

Access Statistics for this article

Journal of Productivity Analysis is currently edited by William Greene, Chris O'Donnell and Victor Podinovski

More articles in Journal of Productivity Analysis from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().