Frontier Estimation and Firm-Specific Inefficiency Measures in the Presence of Heteroscedasticity

Caudill, Steven B.; Ford, Jon M.; Gropper, Daniel M.

doi:10.1080/07350015.1995.10524583

Cited by 251 publications

(173 citation statements)

References 15 publications

Supporting

Mentioning

164

Contrasting

Unclassified

Order By: Relevance

“…The consequences of unconsidered heteroscedasticity on inefficiency estimation (Table 6) have been known in the frontier literature since Caudill et al (1995). Nevertheless, COLS/CML behave rather stably (unlike SF).…”

Section: Misspecificationmentioning

confidence: 95%

See 1 more Smart Citation

Misspecification Preferred: The Sensitivity of Inefficiency Rankings

Jensen

2005

J Prod Anal

View full text Add to dashboard Cite

Ruggiero (European Journal of Operational Research 115, 555-563. 1999) compared the two popular parametric frontier methods for cross-sectional data-the stochastic frontier and the corrected OLS-in a simulation study. He demonstrated that the inefficiency ranking accuracy of the established stochastic frontier is uniformly inferior to that of the misspecified Corrected OLS (COLS) (which lacks an error term). The reason for his result remains unclear, however. In this paper, a more extensive simulation study is therefore conducted to find out whether the superiority of COLS is simply due to small sample sizes or to poor performance of the inefficiency level estimator. JEL Classification: C1, C2, C5

show abstract

Section: Misspecificationmentioning

confidence: 95%

“…• Caudill et al (1995) describe the dramatic consequences of unconsidered heteroscedasticity on inefficiency estimation. Therefore, in the final experiment (Tables 3, 6 and 9), the true inefficiency u * is constructed with multiplicative heteroscedasticity…”

Section: Simulation Designmentioning

confidence: 99%

Misspecification Preferred: The Sensitivity of Inefficiency Rankings

Jensen

2005

J Prod Anal

View full text Add to dashboard Cite

show abstract

“…An example of a model that has the scaling property is the scaled half-normal model, or RSCFG model, of Reifschneider and Stevenson (1991), Caudill and Ford (1993) and Caudill, Ford and Gropper (1995). In this model it is assumed that u it is distributed as N(0, σ 2 it ) + , where σ it (z it , θ)depends in a specific way on z it and some parameters θ .…”

Section: Review Of the Literaturementioning

confidence: 99%

Interpreting and Testing the Scaling Property in Models where Inefficiency Depends on Firm Characteristics

et al. 2006

View full text Add to dashboard Cite

Let u ≥ 0 be technical inefficiency, let z be a set of variables that affect u, and let δ be the parameters of this relationship. The model satisfies the scaling property if u(z, δ) can be written as a scaling function h(z, δ) times a random variable u* that does not depend on z. This property implies that changes in z affect the scale but not the shape of u(z,δ). This paper reviews the existing literature and identifies models that do and do not have the scaling property. It also discusses practical advantages of the scaling property. The paper shows how to test the hypothesis of scaling, and other interesting hypotheses, in the context of the model of Wang, Journal of Productivity Analysis, 2002. Finally, two empirical examples are given. Copyright Springer Science+Business Media, LLC 2006Stochastic frontier model, Scaling property, Technical inefficiency, C12, C31, C52,

show abstract

“…This result is a consequence of using Fare and Primont's (1995b) wrong duality, which results in heteroscedasticity. The form of the heteroscedasticity in the present model is different from that of Caudill, Ford, and Gropper (1995), who assumed an exponential form. Its presence, however, is likely to bias parameter estimates because the expected value of the terms containing e" is nonzero and is dependent on input prices and output quantities.…”

Section: Output Technical Inefficiencymentioning

confidence: 87%

Efficiency measurement with multiple outputs and multiple inputs

Kumbhakar

1996

J Prod Anal

View full text Add to dashboard Cite

AbstmctThis paper discusses modeling technical and allocative inefficiencies in both cost minimizing and profit maximizing frameworks with special emphasis on multiple inputs and multiple outputs. Both primal and dual models are considered for this purpose. In the primal approach we use a separable output and input function (the constant elasticity of transformation output function and Cobb-Douglas input function). The dual models assume translog cost or profti functions. Technical inefficiency is assumed to be random in the cross-sectional models, and fixed firm-specific parameter in the panel data models. Allocative inefficiencies are always treated as input-specific parameters. We derive exact relations linking technical inefficiency and allocative inefficiencies to cost and pm& when the underlying technology is represented by a flexible functional form such as the translog. It is shown that appending a one-sided homoscedastic error term to model technical inefficiency, or neglecting technical in efficiency altogether in a translog pro& tunciton results in model misspecification and inconsistent parameter estimates.

show abstract

Frontier Estimation and Firm-Specific Inefficiency Measures in the Presence of Heteroscedasticity

Cited by 251 publications

References 15 publications

Misspecification Preferred: The Sensitivity of Inefficiency Rankings

Misspecification Preferred: The Sensitivity of Inefficiency Rankings

Interpreting and Testing the Scaling Property in Models where Inefficiency Depends on Firm Characteristics

Efficiency measurement with multiple outputs and multiple inputs

Contact Info

Product

Resources

About