This book includes a spectrum of concepts, such as performance, productivity, operations research, econometrics, and data science, for the practically and theoretically important areas of 'productivity analysis/data envelopment analysis' and 'data science/big data'. Data science is defined as the collection of scientific methods, processes, and systems dedicated to extracting knowledge or insights from data and it develops on concepts from various domains, containing mathematics and statistical methods, operations research, machine learning, computer programming, pattern recognition, and data visualisation, among others. Examples of data science techniques include linear and logistic regressions, decision trees, Naive Bayesian classifier, principal component analysis, neural networks, predictive modelling, deep learning, text analysis, survival analysis, and so on, all of which allow using the data to make more intelligent decisions. On the other hand, it is without a doubt that nowadays the amount of data is exponentially increasing, and analysing large data sets has become a key basis of competition and innovation, underpinning new waves of productivity growth. This book aims to bring a fresh look onto the various ways that data science techniques could unleash value and drive productivity from these mountains of data. Researchers working in productivity analysis/data envelopment analysis will benefit from learning about the tools available in data science/big data that can be used in their current research analyses and endeavours. The data scientists, on the other hand, will also get benefit from learning about the plethora of applications available in productivity analysis/data envelopment analysis.

This book grows from a conference on the state of the art and recent advances in Efficiency and Productivity. Papers were commissioned from leading researchers in the field, and include eight explorations into the analytical foundations of efficiency and productivity analysis. Chapters on modeling advances include reverse directional distance function, a new method for estimating technological production possibilities, a new distance function called a loss distance function, an analysis of productivity and price recovery indices, the relation of technical efficiency measures to productivity measures, the implications for benchmarking and target setting of imposing weight restrictions on DEA models, weight restrictions in a regulatory setting, and the Principle of Least Action. Chapters on empirical applications include a study of innovative firms that use innovation inputs to produce innovation outputs, a study of the impact of potential "coopetition" or cooperation among competitors on the financial performance of European automobile plants, using SFA to estimate the eco-efficiency of dairy farms in Spain, a DEA bankruptcy prediction model, a combined stochastic cost frontier analysis model/mixture hazard model, the evolution of energy intensity in nine Spanish manufacturing industries, and the productivity of US farmers as they age.

This book unifies and extends the definition and measurement of economic efficiency and its use as a real-life benchmarking technique for actual organizations. Analytically, the book relies on the economic theory of duality as guiding framework. Empirically, it shows how the alternative models can be implemented by way of Data Envelopment Analysis. An accompanying software programmed in the open-source Julia language is used to solve the models. The package is a self-contained set of functions that can be used for individual learning and instruction. The source code, associated documentation, and replication notebooks are available online.  The book discusses the concept of economic efficiency at the firm level, comparing observed to optimal economic performance, and its decomposition according to technical and allocative criteria. Depending on the underlying technical efficiency measure, economic efficiency can be decomposed multiplicatively or additively. Part I of the book deals with the classic multiplicative approach that decomposes cost and revenue efficiency based on radial distance functions. Subsequently, the book examines how these partial approaches can be expanded to the notion of profitability efficiency, considering both the input and output dimensions of the firm, and relying on the generalized distance function for the measurement of technical efficiency.  Part II is devoted to the recent additive framework related to the decomposition of economic inefficiency defined in terms of cost, revenue, and profit. The book presents economic models for the Russell and enhanced graph Russell measures, the weighted additive distance function, the directional distance function, the modified directional distance function, and the Hölder distance function. Each model is presented in a separate chapter. New approaches that qualify and generalize previous results are also introduced in the last chapters, including the reverse directional distance function and the general direct approach.  The book concludes by highlighting the importance of benchmarking economic efficiency for all business stakeholders and recalling the main conclusions obtained from many years of research on this topic. The book offers different alternatives to measure economic efficiency based on a set of desirable properties and advises on the choice of specific economic efficiency models.

This book unifies and extends the definition and measurement of economic efficiency and its use as a real-life benchmarking technique for actual organizations. Analytically, the book relies on the economic theory of duality as guiding framework. Empirically, it shows how the alternative models can be implemented by way of Data Envelopment Analysis. An accompanying software programmed in the open-source Julia language is used to solve the models. The package is a self-contained set of functions that can be used for individual learning and instruction. The source code, associated documentation, and replication notebooks are available online. The book discusses the concept of economic efficiency at the firm level, comparing observed to optimal economic performance, and its decomposition according to technical and allocative criteria. Depending on the underlying technical efficiency measure, economic efficiency can be decomposed multiplicatively or additively. Part I of the book deals with the classic multiplicative approach that decomposes cost and revenue efficiency based on radial distance functions. Subsequently, the book examines how these partial approaches can be expanded to the notion of profitability efficiency, considering both the input and output dimensions of the firm, and relying on the generalized distance function for the measurement of technical efficiency. Part II is devoted to the recent additive framework related to the decomposition of economic inefficiency defined in terms of cost, revenue, and profit. The book presents economic models for the Russell and enhanced graph Russell measures, the weighted additive distance function, the directional distance function, the modified directional distance function, and the Hoelder distance function. Each model is presented in a separate chapter. New approaches that qualify and generalize previous results are also introduced in the last chapters, including the reverse directional distance function and the general direct approach. The book concludes by highlighting the importance of benchmarking economic efficiency for all business stakeholders and recalling the main conclusions obtained from many years of research on this topic. The book offers different alternatives to measure economic efficiency based on a set of desirable properties and advises on the choice of specific economic efficiency models.

This book surveys the state-of-the-art in efficiency and productivity analysis, examining advances in the analytical foundations and empirical applications. The analytical techniques developed in this book for efficiency provide alternative ways of defining optimum outcome sets, typically as a (technical) production frontier or as an (economic) cost, revenue or profit frontier, and alternative ways of measuring efficiency relative to an appropriate frontier. Simultaneously, the analytical techniques developed for efficiency analysis extend directly to productivity analysis, thereby providing alternative methods for estimating productivity levels, and productivity change through time or productivity variation across producers. This book includes chapters using data envelopment analysis (DEA) or stochastic frontier analysis (SFA) as quantitative techniques capable of measuring efficiency and productivity. Across the book's 15 chapters, it broadly extends into popular application areas including agriculture, banking and finance, and municipal performance, and relatively new application areas including corporate social responsibility, the value of intangible assets, land consolidation, and the measurement of economic well-being. The chapters also cover topics such as permutation tests for production frontier shifts, new indices of total factor productivity, and also randomized controlled trials and production frontiers.

This book includes a spectrum of concepts, such as performance, productivity, operations research, econometrics, and data science, for the practically and theoretically important areas of 'productivity analysis/data envelopment analysis' and 'data science/big data'. Data science is defined as the collection of scientific methods, processes, and systems dedicated to extracting knowledge or insights from data and it develops on concepts from various domains, containing mathematics and statistical methods, operations research, machine learning, computer programming, pattern recognition, and data visualisation, among others. Examples of data science techniques include linear and logistic regressions, decision trees, Naive Bayesian classifier, principal component analysis, neural networks, predictive modelling, deep learning, text analysis, survival analysis, and so on, all of which allow using the data to make more intelligent decisions. On the other hand, it is without a doubt that nowadays the amount of data is exponentially increasing, and analysing large data sets has become a key basis of competition and innovation, underpinning new waves of productivity growth. This book aims to bring a fresh look onto the various ways that data science techniques could unleash value and drive productivity from these mountains of data. Researchers working in productivity analysis/data envelopment analysis will benefit from learning about the tools available in data science/big data that can be used in their current research analyses and endeavours. The data scientists, on the other hand, will also get benefit from learning about the plethora of applications available in productivity analysis/data envelopment analysis.

This book grows from a conference on the state of the art and recent advances in Efficiency and Productivity. Papers were commissioned from leading researchers in the field, and include eight explorations into the analytical foundations of efficiency and productivity analysis. Chapters on modeling advances include reverse directional distance function, a new method for estimating technological production possibilities, a new distance function called a loss distance function, an analysis of productivity and price recovery indices, the relation of technical efficiency measures to productivity measures, the implications for benchmarking and target setting of imposing weight restrictions on DEA models, weight restrictions in a regulatory setting, and the Principle of Least Action. Chapters on empirical applications include a study of innovative firms that use innovation inputs to produce innovation outputs, a study of the impact of potential "coopetition" or cooperation among competitors on the financial performance of European automobile plants, using SFA to estimate the eco-efficiency of dairy farms in Spain, a DEA bankruptcy prediction model, a combined stochastic cost frontier analysis model/mixture hazard model, the evolution of energy intensity in nine Spanish manufacturing industries, and the productivity of US farmers as they age.