Abstract
Stochastic frontier analysis is a popular tool to assess firm performance. Almost universally it has been applied using maximum likelihood (ML) estimation. An alternative approach, pseudolikelihood (PL) estimation, which decouples estimation of the error component structure and the production frontier, has been adopted in both the non-parametric and panel data settings. To date, no formal comparison has yet to be conducted comparing these methods in a standard, parametric cross-sectional framework. We produce a comparison of these two competing methods using Monte Carlo simulations. Our results indicate that PL estimation enjoys almost identical performance to ML estimation across a range of scenarios and performance metrics, and for certain metrics, outperforms ML estimation when the distribution of inefficiency is incorrectly specified.
| Original language | English (US) |
|---|---|
| Pages (from-to) | 5651-5661 |
| Number of pages | 11 |
| Journal | Applied Economics |
| Volume | 49 |
| Issue number | 55 |
| DOIs | |
| State | Published - Nov 26 2017 |
Fingerprint
Keywords
- maximum likelihood
- Monte Carlo simulation
- production function
- Stochastic frontier analysis
ASJC Scopus subject areas
- Economics and Econometrics
Cite this
Pseudolikelihood estimation of the stochastic frontier model. / Andor, Mark; Parmeter, Christopher.
In: Applied Economics, Vol. 49, No. 55, 26.11.2017, p. 5651-5661.Research output: Contribution to journal › Article
}
TY - JOUR
T1 - Pseudolikelihood estimation of the stochastic frontier model
AU - Andor, Mark
AU - Parmeter, Christopher
PY - 2017/11/26
Y1 - 2017/11/26
N2 - Stochastic frontier analysis is a popular tool to assess firm performance. Almost universally it has been applied using maximum likelihood (ML) estimation. An alternative approach, pseudolikelihood (PL) estimation, which decouples estimation of the error component structure and the production frontier, has been adopted in both the non-parametric and panel data settings. To date, no formal comparison has yet to be conducted comparing these methods in a standard, parametric cross-sectional framework. We produce a comparison of these two competing methods using Monte Carlo simulations. Our results indicate that PL estimation enjoys almost identical performance to ML estimation across a range of scenarios and performance metrics, and for certain metrics, outperforms ML estimation when the distribution of inefficiency is incorrectly specified.
AB - Stochastic frontier analysis is a popular tool to assess firm performance. Almost universally it has been applied using maximum likelihood (ML) estimation. An alternative approach, pseudolikelihood (PL) estimation, which decouples estimation of the error component structure and the production frontier, has been adopted in both the non-parametric and panel data settings. To date, no formal comparison has yet to be conducted comparing these methods in a standard, parametric cross-sectional framework. We produce a comparison of these two competing methods using Monte Carlo simulations. Our results indicate that PL estimation enjoys almost identical performance to ML estimation across a range of scenarios and performance metrics, and for certain metrics, outperforms ML estimation when the distribution of inefficiency is incorrectly specified.
KW - maximum likelihood
KW - Monte Carlo simulation
KW - production function
KW - Stochastic frontier analysis
UR - http://www.scopus.com/inward/record.url?scp=85019729222&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85019729222&partnerID=8YFLogxK
U2 - 10.1080/00036846.2017.1324611
DO - 10.1080/00036846.2017.1324611
M3 - Article
AN - SCOPUS:85019729222
VL - 49
SP - 5651
EP - 5661
JO - Applied Economics
JF - Applied Economics
SN - 0003-6846
IS - 55
ER -