# Managerial Performance Analysis Using R

### Venue

## About the Course

This three-day intensive course presents a new and coherent framework for analysing the performance of firm managers. The focus is on measures of performance that are of practical relevance to policy makers. Most, if not all, of these measures can be viewed as measures of productivity and/or efficiency. Unlike many courses in managerial performance analysis, participants will learn how to measure productivity in ways that are consistent with measurement theory. They will also learn how piecewise frontier models (PFMs), deterministic frontier models (DFMs) and stochastic frontier models (SFMs) can be used to explain variations in productivity. They will obtain hands-on experience implementing these methods using R. R is a free open-source software package that is currently revolutionizing statistical analysis in all branches of academia, government and industry.

This course is supported by AIBE

## About the Presenter

Chris O’Donnell is a Professor of Econometrics at the University of Queensland. He is a Co-editor of the Journal of Productivity Analysis, an Associate Editor of Empirical Economics, and a Distinguished Fellow of the Australian Agricultural and Resource Economics Society.

His research interests focus on mathematical programming and statistical methods for measuring and analysing productivity and efficiency change. His work has been published in leading economics and econometrics journals, including the American Journal of Agricultural Economics, the Journal of Econometrics, the Journal of Applied Econometrics, and Econometric Reviews. He has provided in-house training and/or been a consultant for organisations including the World Bank, the Asian Productivity Organisation, the International Rice Research Institute, the Australian Energy Regulator, the New South Wales Independent Pricing and Regulatory Tribunal, and the Australian Independent Hospital Pricing Authority.

## Who Should Attend?

The course is aimed at graduate students, researchers, economists, statisticians and consultants from private and public sector organizations, regulatory authorities, regulated firms, infrastructure industries (e.g., electricity, gas, railways), service industries (e.g., education, health), and industries with branch structures (e.g., banks, credit unions, franchises, retail chains). Participants are expected to have an understanding of microeconomics and econometrics similar to that of a second-year undergraduate economics student at an Australian university

## Course Outline

Course instruction will take the form of lectures and tutorial sessions. The tutorial sessions will give participants hands-on experience using R. The course will cover the following topics/modules:

**Production Technologies. **

To analyse managerial performance, it is necessary to know something about what can and cannot be produced using different production technologies. The input-output combinations that are possible using different technologies in different environments can be represented by output sets, input sets and production possibilities sets. Under certain conditions, they can also be represented by distance, revenue, cost and profit functions. This topic defines, and discusses the properties of, these different sets and functions.

**Managerial Behaviour.**

To analyse managerial performance, it is necessary to know or assume something about managerial behaviour. The existence of cost functions, for example, does not mean that managers will aim to minimise costs. Rather, different managers will tend to behave differently depending on what they value and what they can and cannot choose. For example, if managers value goods and services at market prices, then, if possible, they will tend to choose outputs and inputs to maximise profits. This topic discusses some of the simplest optimisation problems faced by firm managers.

**Measures of Efficiency.**

In this course, measures of efficiency are viewed as *ex post *measures of how well firm managers have solved different optimisation problems. For example, measures of profit efficiency are viewed as measures of how well managers have maximised profits when inputs and outputs have been chosen freely. This topic discusses various output-, input-, revenue-, cost-, profit- and productivity-oriented measures of efficiency.

**Index Numbers. **

An index is a rule or formula that explains how to use data to measure the change in one or more variables across time and/or space. This topic discusses output quantity, input quantity and productivity indices. The focus is on indices that are proper in the sense that they satisfy a set of important axioms from index theory. One of these axioms is transitivity. If firm A produced twice as much as firm B, and firm B produced twice as much as firm C, then a transitive index would say that firm A produced four times as much as firm C. The class of proper indices includes various additive, multiplicative, primal and dual indices. Indices that are NOT proper include the widely-used Fisher, Tornqvist, Malmquist, EKS and CCD indices.

**Piecewise Frontier Models.**

Explaining changes in managerial performance involves estimating changes in production frontiers. A widely-used estimation approach involves enveloping scatterplots of data points as tightly as possible without violating any assumed characteristics of production technologies. Some of the most common assumptions lead to estimated frontiers that are comprised of multiple linear segments (or pieces). The associated frontiers are known as piecewise frontiers. This topic explains how to estimate the unknown parameters of so-called piecewise frontier models (PFMs). It then explains how the estimated parameters can be used to analyse managerial performance. The focus is on what are commonly known as data envelopment analysis (DEA) estimators.

**Deterministic Frontier Models.**** **

Production frontiers are often represented by parametric functions. These functions can be written in the form of regression models in which the explanatory variables are often deterministic (i.e., not random). This topic explains how to estimate and draw inferences concerning the unknown parameters of so-called deterministic frontier models (DFMs). It then explains how the estimated parameters can be used to analyse managerial performance. The focus is on least squares (LS) estimators.

**Stochastic Frontier Models.**** **

Distance, revenue, cost and profit functions can be written in the form of regression models with unobserved error terms representing statistical noise and different types of inefficiency. In practice, the noise components are almost always assumed to be random variables (i.e., stochastic). The associated frontiers are known as stochastic frontiers. This topic explains how to estimate and draw inferences concerning the unknown parameters of so-called stochastic frontier models (SFMs). It then explains how the estimated parameters can be used to analyse managerial performance. The focus is on maximum likelihood (ML) estimators.

**Practical Issues.**** **

Policy-oriented performance analysis involves a number of steps that are best completed in a prescribed order or sequence. For example, it is usually best to fully describe the production unit before attempting to list and classify all the variables that are physically involved in the production process. It is generally also best to identify relevant measures of performance before assembling the data and choosing an estimation approach. This topic describes the main steps involved in policy-oriented performance analysis. It also discusses some policy options for targeting the different drivers of managerial performance.

Standard Delegate: $1000

Full-Time (FT) Student: $500

UQ Staff Member: $700^

UQ FT Student: $350^

^Email event contact to process fee