Indicators of ill-conditioned data sets and model misspecification in Data Envelopment Analysis: an extended facet approach

Article Abstract:

Data Envelopment Analysis (DEA) measures the relative efficiency of Decision Making Units through the use of mathematical programming. A study is conducted with the aim of developing indicators of ill-conditioned data sets and model misspecification in the DEA approach. More specifically, it strives to look for determinants of whether or not data support the specification of the input and output space, that is, if the variation in data is enough to estimate a frontier of the same dimension as the input/output space. It is found that the presence of well-defined marginal rates of substitution along the estimated strongly efficient frontier segments calls for the existence of Full Dimension Efficient Facets (FDEFs). A test for the presence of FDEFs is created and an operational two-stage methodology for evaluating efficiency is presented.

---

Chance constrained efficiency evaluation

Article Abstract:

A model for efficiency evaluation founded on the theory of chance constrained programming is constructed. This model employs a piece-by-piece linear envelopment of confidence areas for observed stochastic multiple-input multiple-output mixtures in a mathematical programming approach, in the tradition of the Data Envelopment Analysis (DEA). An exogenous decomposition of the full variation in data for each Decision Making Unit (DMU) is allowed. The model can generate estimates of the sensitivity of efficiency scores to unknown levels of data noise by changing some probability levels. An economic interpretation of the suggested efficiency measure which is similar to the interpretation of the CCR-efficiency index is possible. The model is applied to research evaluations in economic departments at Danish Universities.

Models, Industrial efficiency, Economic efficiency

---

Simulating weights restrictions in data envelopment analysis by means of unobserved DMUs

Article Abstract:

A study was conducted to characterize a method of obtaining and utilizing value judgments in data envelopment analysis (DEA), a technique for evaluating the comparative efficiencies of decision making units (DMU) that come in a range of scales of operation and weights restrictions. Based on the use of unobserved decision making units, the method promotes local, nonlinear marginal rates of substitution between inputs and outputs. Capturing value judgments in DEA using unobserved DMUs presents several benefits, including the creation of more than one unobserved DMU from a given real DMU, support for radial measure of efficiency and the ability to obtain more value judgments in DEA under variable returns to scale.

Decision-making, Decision making, Linear programming, Usage

Similar abstracts:

• Abstracts: Agency costs, institutions, learning, and taxation in venture capital contracting. Crowding out private equity: Canadian evidence
• Abstracts: Core objectives: clarity in designing strategy. Architecture and communication among product development engineers
• Abstracts: The effects of time pressure and knowledge on key word selection behavior in tax research. Do effects of client preference on accounting professionals' information search and subsequent judgments persist with high practice risk?
• Abstracts: Contests can rev up employees: additional incentives for salespeople and other workers can generate excitement and boost profits
• Abstracts: An empirical investigation of the linkages among relationalism, environmental uncertainty, and bureaucratization

This website is not affiliated with document authors or copyright owners. This page is provided for informational purposes only. Unintentional errors are possible.