Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 11 of the 11 references it cites were found in a registry or an index; none of the 12 sources its claims cite is checked for support yet.

One-Step and Two-Step Estimation of the Effects of Exogenous Variables on Technical Efficiency Levels

Open access
Wang, Hung-jen, Schmidt, Peter · 2002 · Journal of Productivity Analysis

Overview

CoChat Summary

## TL;DR

This paper compares one-step and two-step estimation for stochastic frontier models when exogenous variables affect technical inefficiency. In Monte Carlo simulations, the one-step estimator performs well, while the two-step procedure shows serious bias in both the first and second stages.

## Key findings

  • In the base case simulation (200 replications, sample size \(N=200\)), the one-step estimator is close to unbiased: \(\hat\beta=-0.0010\), \(\hat\delta=1.0010\), and average estimated efficiency \(E(\hat e^{-u})=0.5141\) versus the true \(0.5165\).
  • Under the same base case, the two-step estimator is badly biased: \(\hat\beta=-0.3064\), estimated efficiency \(E(\hat e^{-u})=0.4482\) versus true \(0.52\) (table reports \(0.5165\) in the design and about \(0.52\) in results discussion), and the second-step estimate of \(\delta\) is \(0.3539\) versus the true value \(1\).
  • The paper reports that the two-step bias persists even when \(x\) and \(z\) are independent: for \(\rho=0\), the second-step estimate of \(\delta\) is still \(0.4069\), and the estimated efficiencies remain underdispersed and downward biased.
  • Increasing sample size from \(N=200\) to \(N=500\) and \(N=1000\) lowers standard deviations but does not remove the two-step biases; for example, the two-step \(\hat\delta\) stays around \(0.35\)–\(0.36\) while the true value is \(1\).

## Methods

  • The paper studies a stochastic frontier model in which inefficiency depends on exogenous variables \(z\), emphasizing a “scaling property” specification where the one-sided inefficiency term is a function of \(z\) times a basic one-sided error.
  • It uses Monte Carlo simulations with 2,000 replications per design point. The base case sets \(\beta=0\), \(\delta=1\), \(\rho=0.5\), \(\gamma=1\), \(\sigma_v=1\), \(\mu=0\), and \(N=200\).
  • One-step estimation is maximum likelihood under the correctly specified model; two-step estimation first sets \(\delta=0\) in the frontier MLE, then regresses \(\ln(\hat u)\) on \(z\) to estimate \(\delta\).

## Limitations

  • The paper’s own simulation evidence is limited to one simple model with the scaling property and a specific truncated-normal basic distribution; results may not generalize beyond that setup.
  • The text makes evident, though it does not frame it as a formal limitation, that some parameters (\(\mu\) and \(\gamma\)) are “hard” to estimate and produced outliers severe enough that the authors truncated the most extreme 0.3% of replications for those parameters.
  • Not stated in the paper.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1023/a:1016565719882

Cites

11 of 11 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Verification

Claims the paper makes

0 Valid · 4 Not valid · 8 Not checked · 112 Own results and interpretations

63 more claims

Create a free account to see all 124 claims and every source verdict.

See all claims — free account
  • It may state two separate claims as one. 43
  • Probably not a substantial claim. 16
  • The quote only partly states this claim. 5

Create a free account to see these claims.

Citations

Cites

Show all 10 references

Cited by

See all 1,056

Metadata

Authors
Wang, H.-J., & Schmidt, P.
Affiliations
  • Institute of Economics, Academia Sinica (TW)
  • Michigan State University (US)
Venue
Journal of Productivity Analysis
Published
2002-09
Volume
18
Issue
2
Pages
129-144
ISSN
  • 0895-562X (ISSN-L)
  • 1573-0441
OpenAlex
W1524136776
Cited by
1,045
References
10
Citations by year
2026: 42 · 2025: 56 · 2024: 63 · 2023: 65 · 2022: 61 · 2021: 74 · 2020: 63 · 2019: 64 · 2018: 70 · 2017: 51 · 2016: 48 · 2015: 55 · 2014: 49 · 2013: 39 · 2012: 34
FWCI
17.56
Type
journal article
Language
en
Open access
Licence
Locations
  • Journal of Productivity Analysis (publishedVersion)
  • Munich Personal RePEc Archive (Ludwig Maximilian University of Munich) (submittedVersion)
  • RePEc: Research Papers in Economics (submittedVersion)
Topics
  • Social Sciences › Decision Sciences › Management Science and Operations Research › Efficiency Analysis Using DEA
  • Social Sciences › Economics, Econometrics and Finance › Economics and Econometrics › Economic Growth and Productivity
  • Social Sciences › Economics, Econometrics and Finance › Economics and Econometrics › Spatial and Panel Data Analysis
Keywords
stochastic frontier analysis, technical efficiency, firm characteristics, one-step estimation, maximum likelihood estimation, Monte Carlo simulation

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

A language model drew the claims from the paper on 2026-10-11, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-11 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: One-Step and Two-Step Estimation of the Effects of Exogenous Variables on Technical Efficiency Levels. https://app.cochat.ai/papers/10.1023/a:1016565719882
The paper
Wang, H.-J., & Schmidt, P. (2002). One-Step and Two-Step Estimation of the Effects of Exogenous Variables on Technical Efficiency Levels. Journal of Productivity Analysis. https://doi.org/10.1023/a%3A1016565719882

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.