Spelling suggestions: "subject:"frontier anestimation"" "subject:"frontier 3destimation""
1 |
Essays on Regression Spline Structural Nonparametric Stochastic Production Frontier Estimation and Inefficiency Analysis ModelsLi, Ke 2010 December 1900 (has links)
Conventional Cobb-Douglas and Transcendental Logarithmic production functions widely used in Stochastic Production Frontier Estimation and Inefficiency Analysis have merits and deficiencies. The Cobb-Douglas function imposes monotonicity and concavity constraints required by microeconomic theory. However it is inflexible and implies undesired assumptions as well. The Trans-log function is very flexible and does not imply undesired assumptions, yet it is very hard to impose both monotonicity and concavity constraints. The first essay introduced a class of stochastic production frontier estimation models that impose monotonicity and concavity constraints and suggested models that are very flexible. Researchers can use arbitrary order of polynomial functions or any function of independent variables within the suggested frameworks. Also shown was that adopting suggested models could greatly increase predictive accuracy through simulations. In the second essay we generalized the suggested models with the Inefficiency Analysis technique. In the last essay we extended the models developed in the previous two essays with regression spline and let the data decide how flexible or complicated the model should be. We showed the improvement of deterministic frontier estimation this extension could bring through simulations, as well. Works in this dissertation reduced the gap between conventional structural models and nonparametric models in stochastic frontier estimation field. This dissertation offered applied researchers Stochastic Production Frontier models that are more accurate and flexible than previous ones. It also preserves constraints of economic theory.
|
2 |
A new approach to stochastic frontier estimation: DEA+Gstach, Dieter January 1996 (has links) (PDF)
The outcome of a production process might not only deviate from a theoretical maximum due to inefficiency, but also because of non-controllable influences. This raises the issue of reliability of Data Envelopment Analysis in noisy environments. I propose to assume an i.i.d. data generating process with bounded noise component, so that the following approach is feasible: Use DEA to estimate a pseudo frontier first (nonparametric shape estimation). Next apply a ML-technique to the DEA-estimated efficiencies, to estimate the scalar value by which this pseudo-frontier must be shifted downward to get the true production frontier (location estimation). I prove, that this approach yields consistent estimates of the true frontier. (author's abstract) / Series: Department of Economics Working Paper Series
|
3 |
Production Economics Modeling and Analysis of Polluting firms: The Production Frontier ApproachMekaroonreung, Maethee 2012 August 1900 (has links)
As concern grows about energy and environment issues, energy and environmental modeling and related policy analysis are critical issues for today's society. Polluting firms such as coal power plants play an important role in providing electricity to drive the U.S. economy as well as producing pollution that damages the environment and human health. This dissertation is intended to model and estimate polluting firms' production using nonparametric methods. First, frontier production function of polluting firms is characterized by weak disposability between outputs and pollutants to reflecting the opportunity cost to reduce pollutants. The StoNED method is extended to estimate a weak disposability frontier production function accounting for random noise in the data. The method is applied to the U.S. coal power plants under the Acid Rain Program to find the average technical inefficiency and shadow price of SO2 and NOx. Second, polluting firms' production processes are modeled characterizing both the output production process and the pollution abatement process. Using the law of conservation of mass applied to the pollution abatement process, this dissertation develops a new frontier pollutant function which then is used to find corresponding marginal abatement cost of pollutants. The StoNEZD method is applied to estimate a frontier pollutant function considering the vintage of capital owned by the polluting firms. The method is applied to estimate the average NOx marginal abatement cost for the U.S. coal power plants under the current Clean Air Interstate Rule NOx program. Last, the effect of a technical change on marginal abatement costs are investigated using an index decomposition technique. The StoNEZD method is extended to estimate sequential frontier pollutant functions reflecting the innovation in pollution reduction. The method is then applied to estimate a technical change effect on a marginal abatement cost of the U.S. coal power plants under the current Clean Air Interstate Rule NOx program.
|
4 |
Small sample performance of two approaches to technical efficiency estimation in noisy multiple output environmentsGstach, Dieter January 1998 (has links) (PDF)
This paper provides simulation evidence concerning some statistical properties of two different approaches to technical efficiency estimation for multiple-output production under noisy conditions: The Ray Frontier Approach (RFA) from Löthgren (1997) DEA+ proposed in Gstach (1996). RFA, unlike earlier approaches in the realm of stochastic frontier analysis, is capable of efficiency estimation in the case of multiple outputs as well and lends itself for comparison with DEA+. Several settings with varying sample sizes, noise to signal ratios and mean inefficiencies are investigated. (author's abstract) / Series: Department of Economics Working Paper Series
|
5 |
Essays on the effect of product recalls and mergers on firm performanceYan, Wenfeng 21 March 2012 (has links)
This dissertation investigates the effect of endogenous and exogenous events on firm behavior and performance. These are fundamental questions in economics. The contribution of this study is threefold. First, it provides estimates of the impact of mergers on railroad efficiency, which has important antitrust implications. Second, it provides new estimates of the effect of negative events on the market value of Johnson & Johnson, Bridgestone, and Toyota, which is important to the understanding of how markets punish corporate errors. Third, it develops better ways to estimate these effects.
Chapter 2 uses the event study approach to determine how product recalls due to exogenous and endogenous shocks affect the value of the firm. Three recalls from Johnson & Johnson, Bridgestone, and Toyota have been studied in this chapter. The traditional event study method assumes that markets are efficient, a questionable assumption in the short run. Thus, the current stock value of a firm may not reflect its true market value. To address this potential problem, frontier based methods are used, including data envelopment analysis, corrected ordinary least squares, and stochastic frontier regression analysis. Stochastic frontier methods are shown to be more appropriate when market behavior is not fully rational. The evidence shows that endogenous events due to firm errors are more detrimental to firm value than exogenous negative events that are beyond the control of the firm. That is, the market is more forgiving of negative shocks that the company cannot control.
Chapter 3 studies the effects of merger activity on the efficiency and productivity growth of U.S. Class I railroads from 1983 to 2008. In this chapter, I assess the effects of merger activity on efficiency, and identify the major factors associated with productivity growth. Unlike previous research, I use data envelopment analysis with an attribute-incorporated Malmquist productivity index. This approach allows firm specific measures of efficiency and productivity to be calculated for firms with differences in technology. The approach allows a decomposition of the attribute-incorporated Malmquist productivity index into technical, efficient and attribute components, the impacts of railroads mergers, and the real source and change of productivity. I find that (1) the technology efficiency performance of the seven survivor firms has grown through time; (2) mergers overall do not lead significant technology and scale efficiency gains, but there are differences across mergers; (3) mergers in the 1980s do not have significant different effect on efficiency change compared to those in the 1990s; and (4) the productivity gains are mostly attributed to the network and operation attributes change and industry technology improvement. Overall, the mergers have no direct impact on the efficiency gains or losses during our study period.
The application of these techniques to product recalls and railroad merger models demonstrates how they can provide superior estimates over traditional estimation techniques. It is hoped that these applications will motivate the use of these techniques in other settings. / Graduation date: 2012
|
Page generated in 0.0821 seconds