Autoregressive stochastic frontier model with nonlinear function of exogenous variable and dynamic technical inefficiency

Subject Areas : Statistics

Bahareh Feizi ¹ , Ahmad Poyrdarvish ²

1 - Department of Mathematics, Chalous Branch, Islamic Azad University, Chalous, Iran
2 - Faculty of Mathematical Sciences, University of Mazandaran, Babolsar, Iran

Received: 2021-10-10 Accepted : 2022-03-14 Published : 2022-12-22

Keywords: حداکثر انتظار, برآورد نیمه‌پارامتری, مدل مرزی تصادفی خودبازگشتی, داده‌های تابلویی, ناکارایی فنی,

Abstract :

In this paper, we focus on a new autoregressive stochastic frontier model based on the nonlinear function of the exogenous variable in the panel data. In classical financial models, the technical inefficiency is considered uncorrelated, which is often not satisfied in empirical cases. In the proposed autoregressive stochastic frontier model, the error consists of two components, statistical error, and technical inefficiency, so that technical inefficiency is assumed to be autocorrelated. Autocorrelated technical inefficiency can be interpreted as the technical inefficiency of a company at the current time depends on the extent of the company's previous technical inefficiency and its current transient inefficiency. The semiparametric method for estimating the nonlinear function of the exogenous variable is calculated through a two-step process with Taylor series expansion and nonparametric adjustment factor. The expectation-maximization approach is used to estimate the parameters of the model and the estimation performance is evaluated by Monte Carlo simulation.

References:

[1] D.J. Aigner, C.A.K. Lovell, P. Schmidt. Formulation and estimation of stochastic frontier production function models. Journal of Econometrics 6:21–37(1977)

 

[2] W. Meeusen, J. van den Broeck. Efficiency estimation from Cobb-Douglas production functions with composed error. International Economic Review 18:435–444(1977)

 

[3] M. Koetter, J.W. Kolari, L. Spierdijk. Enjoying the quiet life under deregulation? evidence from adjusted Lerner indices for U.S. banks. The Review of Economics and Statistics 94:462–480(2012)

 

[4] Casu, B., Ferrari, A. and Zhao, T. Regulatory reform and productivity change in Indian banking. Review of Economics and Statistics, 95(3):1066–1077(2013)

 

[5] Baten, A. and Hossain, I. Stochastic frontier model with distributional assumptions for rice production technical efficiency. Journal of Agricultural Science and Technology, 16:(2014)

 

[6] M.M. Pitt, L.-F. Lee. The measurement and sources of technical inefficiency in the Indonesian weaving industry. Journal of Development Economics 9:43–64(1981)

 

[7] P. Schmidt, R.C.  Sickles. Production frontiers and panel data. Journal of Business and Economic Statistics 2:367–374(1984)

 

[8] H. Lai, S.C. Kumbhakar. Estimation of dynamic stochastic frontier model using likelihood-based approaches. MPRA Paper, 87830, University Library of Munich, Germany (2018)

 

[9] B. Feizi, A. Pourdarvish. Nonlinear autoregressive stochastic frontier model with dynamic technical inefficiency in panel data. Communications in Statistics - Theory and Methods, (2021)

 

[10] H. Noh, I.V. Keilegom. On relaxing the distributional assumption of stochastic frontier models. Journal of the Korean Statistical Society 49:1–14(2020)

 

[11] Alsumaiei, A.A. A Nonlinear Autoregressive modeling approach for forecasting groundwater level fluctuation in urban aquifers. Water, 12: 820(2020)

 

[12] A. Tealab, H. Hefny, A. Badr. Forecasting of nonlinear time series using ANN. Future Computing and Informatics Journal 2(1):39–47(2017)

 

[13] S. Paul, S. Shankar. Estimating efficiency effects in a panel data stochastic frontier model. Journal of Productivity Analysis 53:163–180(2020)

 

[14] Yu, Z., Wang, b. and Shi, N. Semiparametric estimation of regression functions in autoregressive models. Statistics and Probability Letters, 79:165–172(2009)

 

[15] R. Farnoosh, M. Hajebi, S.Y. Samadi. A semiparametric estimation for the first-order nonlinear autoregressive time series model with independent and dependent errors. Iranian Journal of Science and Technology 43(3):905–917(2019)

 

[16] Azzalini A. with the collaboration of Capitanio A. The skew-normal and related families. Cambridge: Cambridge University Press, IMS Monographs series (2014)

 

[17] M. Pourahmadi. Skew-normal ARMA models with nonlinear heteroscedastic predictors. Communications in Statistics - Theory and Methods 36:1803–1819(2007)

 

[18] M. Sharafi, and A.R. Nematollahi. AR(1) model with skew-normal innovations. Metrika 17:1011–1029(2016)

 

[19] M. Maleki. R.B.  Arellano-Valle. Maximum a-posteriori estimation of autoregressive processes based on finite mixtures of scale-mixtures of skew-normal distributions. Journal of Statistical Computation and Simulation 87(6):1061–1083(2017)

 

[20] A. Hajrajabi, A. Fallah. Nonlinear semiparametric AR(1) model with skewsymmetric innovations. Communications in Statistics - Simulation and Computation 47(5):1453–1462(2018)