Estimating stochastic production frontiers: A one-stage multivariate semiparametric Bayesian concave regression method

This paper describes a method to estimate a production frontier that satisfies the axioms of monotonicity and concavity in a non-parametric Bayesian setting. An inefficiency term that allows for significant departure from prior distributional assumptions is jointly estimated in a single stage with parametric prior assumptions. We introduce heteroscedasticity into the inefficiency terms by local hyperplane-specific shrinkage hyperparameters and impose monotonicity using bound-constrained local nonlinear regression. Our minimum-of-hyperplanes estimator imposes concavity. Our Monte Carlo simulation experiments demonstrate that the frontier and efficiency estimations are competitive, economically sound, and allow for the analysis of larger datasets than existing nonparametric methods. We validate the proposed method using data from 2007-2010 for Japan's concrete industry. The results show that the efficiency levels remain relatively high over the time period.

[1]  Gareth O. Roberts,et al.  On convergence of the EM algorithmand the Gibbs sampler , 1999, Stat. Comput..

[2]  Jeffrey S. Racine,et al.  Smooth constrained frontier analysis , 2013 .

[3]  Thomas S. Shively,et al.  Nonparametric function estimation subject to monotonicity, convexity and other shape constraints , 2011 .

[4]  M. Steel,et al.  Semiparametric Bayesian Inference for Stochastic Frontier Models , 2004 .

[5]  Léopold Simar,et al.  A general framework for frontier estimation with panel data , 1996 .

[6]  Timo Kuosmanen,et al.  One-stage estimation of the effects of operational conditions and practices on productive performance: asymptotically normal and efficient, root-n consistent StoNEZD method , 2011 .

[7]  Stefan Seifert,et al.  Environmental factors in frontier estimation - A Monte Carlo analysis , 2018, Eur. J. Oper. Res..

[8]  Rolf Färe,et al.  On piecewise reference technologies , 1988 .

[9]  Adrian F. M. Smith,et al.  Automatic Bayesian curve fitting , 1998 .

[10]  Gad Allon,et al.  Nonparametric Estimation of Concave Production Technologies by Entropic Methods , 2005 .

[11]  A. U.S.,et al.  FORMULATION AND ESTIMATION OF STOCHASTIC FRONTIER PRODUCTION FUNCTION MODELS , 2001 .

[12]  Chad Syverson,et al.  Market Structure and Productivity: A Concrete Example , 2004, Journal of Political Economy.

[13]  Andrew L. Johnson,et al.  Stochastic Nonparametric Approach to Efficiency Analysis: A Unified Framework , 2015 .

[14]  Andrew L. Johnson.  STOCHASTIC DEA: THE REGRESSION-BASED APPROACH , 2017 .

[15]  Léopold Simar,et al.  Frontier estimation in the presence of measurement error with unknown variance , 2015 .

[16]  Andrew Gelman,et al.  Data Analysis Using Regression and Multilevel/Hierarchical Models , 2006 .

[17]  Robin C. Sickles,et al.  Semiparametric Estimation under Shape Constraints , 2017 .

[18]  Jennifer A. Hoeting,et al.  Bayesian estimation and inference for generalised partial linear models using shape-restricted splines , 2011 .

[19]  Timo Kuosmanen,et al.  A more efficient algorithm for Convex Nonparametric Least Squares , 2013, Eur. J. Oper. Res..

[20]  W. Griffiths,et al.  Estimating State‐Contingent Production Frontiers , 2006 .

[21]  P. W. Wilson,et al.  Inferences from Cross-Sectional, Stochastic Frontier Models , 2009 .

[22]  M. Steel,et al.  Stochastic frontier models: a bayesian perspective , 1994 .

[23]  M. Zyphur,et al.  Bayesian Estimation and Inference , 2015 .

[24]  P. Hall,et al.  Estimating a Changepoint, Boundary, or Frontier in the Presence of Observation Error , 2002 .

[25]  Yanqin Fan,et al.  Semiparametric Estimation of Stochastic Production Frontier Models , 1996 .

[26]  Levent Kutlu.  Battese-coelli estimator with endogenous regressors , 2010 .

[27]  L. Simar,et al.  Stochastic FDH/DEA estimators for frontier analysis , 2008 .

[28]  D. Poirier,et al.  Bayesian Variants of Some Classical Semiparametric Regression Techniques , 2004 .

[29]  Timothy Coelli,et al.  A Bayesian Approach to Imposing Curvature on Distance Functions , 2005 .

[30]  Timo Kuosmanen.  Representation Theorem for Convex Nonparametric Least Squares , 2008, Econometrics Journal.

[31]  D. Dunson,et al.  Bayesian nonparametric multivariate convex regression , 2011, 1109.0322.

[32]  C. Lovell,et al.  On the estimation of technical inefficiency in the stochastic frontier production function model , 1982 .

[33]  E. Tsionas,et al.  Bayesian Treatments for Panel Data Stochastic Frontier Models with Time Varying Heterogeneity , 2017 .

[34]  Timo Kuosmanen,et al.  Stochastic non-smooth envelopment of data: semi-parametric frontier estimation subject to shape constraints , 2012 .

[35]  Panayotis G. Michaelides,et al.  Global approximation to arbitrary cost functions: A Bayesian approach with application to US banking , 2015, Eur. J. Oper. Res..

[36]  W. Griffiths,et al.  Some models for stochastic frontiers with endogeneity , 2016 .

[37]  Léopold Simar,et al.  Pitfalls of Normal-Gamma Stochastic Frontier Models , 1997 .

[38]  W. Diewert,et al.  Flexible Functional Forms and Global Curvature Conditions , 1989 .

[39]  Kien C. Tran,et al.  GMM estimation of stochastic frontier model with endogenous regressors , 2013 .

[40]  Léopold Simar,et al.  Nonparametric stochastic frontiers: A local maximum likelihood approach , 2007 .

[41]  Peter J. Klenow,et al.  Misallocation and Manufacturing TFP in China and India , 2007 .

[42]  Andrew L. Johnson,et al.  The Effect of Performance Measurement Systems on Productive Performance: An Empirical Study of Italian Manufacturing Firms , 2015 .

[43]  P. Hall,et al.  NONPARAMETRIC KERNEL REGRESSION SUBJECT TO MONOTONICITY CONSTRAINTS , 2001 .

[44]  Mike G. Tsionas,et al.  Zero-inefficiency stochastic frontier models with varying mixing proportion: A semiparametric approach , 2016, Eur. J. Oper. Res..

[45]  R. Banker,et al.  Maximum likelihood estimation of monotone and concave production frontiers , 1992 .

[46]  Nicolas Chopin,et al.  Fast simulation of truncated Gaussian distributions , 2011, Stat. Comput..

[47]  C. Hildreth.  Point Estimates of Ordinates of Concave Functions , 1954 .

[48]  Badi H. Baltagi,et al.  A General Index of Technical Change , 1988, Journal of Political Economy.

[49]  P. Green.  Reversible jump Markov chain Monte Carlo computation and Bayesian model determination , 1995 .

[50]  K. Nakata,et al.  Sectoral Productivity and Economic Growth in Japan: 1970-98 , 2003 .

[51]  Chad Syverson,et al.  Product Substitutability and Productivity Dispersion , 2003, Review of Economics and Statistics.

[52]  Christopher F. Parmeter,et al.  Imposing Economic Constraints in Nonparametric Regression: Survey, Implementation and Extension , 2009, SSRN Electronic Journal.

[53]  Xiaofeng Dai.  Non-parametric efficiency estimation using Richardson-Lucy blind deconvolution , 2016, Eur. J. Oper. Res..

[54]  David B. Dunson,et al.  Multivariate convex regression with adaptive partitioning , 2011, J. Mach. Learn. Res..

[55]  Robin C. Sickles,et al.  The Skewness Issue in Stochastic Frontiers Models: Fact or Fiction? , 2011 .

[56]  David B. Dunson,et al.  Ensemble Methods for Convex Regression with Applications to Geometric Programming Based Circuit Design , 2012, ICML.

[57]  Daniel A. Ackerberg,et al.  Identification Properties of Recent Production Function Estimators , 2015 .

[58]  Jeffrey S. Racine,et al.  Nonparametric Kernel Regression with Multiple Predictors and Multiple Shape Constraints , 2012 .

[59]  Timo Kuosmanen,et al.  Data Envelopment Analysis as Nonparametric Least-Squares Regression , 2010, Oper. Res..

[60]  Efthymios G. Tsionas,et al.  Full Likelihood Inference in Normal-Gamma Stochastic Frontier Models , 2000 .