Ranking Universities using Data Envelopment Analysis

Similar documents
Evaluating Iran SME s R&D Efficiency Provinces using DEA

Measuring Efficiency of Foreign Banks in the United States

Technical Efficiency of Management wise Schools in Secondary School Examinations of Andhra Pradesh by CCR Model

A Study of the Efficiency of Polish Foundries Using Data Envelopment Analysis

Blessing or Curse from Health Insurers Mergers and Acquisitions? The Analysis of Group Affiliation, Scale of Operations, and Economic Efficiency

Efficiency Measurement of Turkish Public Universities with Data Envelopment Analysis (DEA)

Allocation of shared costs among decision making units: a DEA approach

Organised by the Croatian National Bank. Boris Vujčić Igor Jemrić. Efficiency of Banks in Transition: A DEA Approach

Measuring the Efficiency of Public Transport Sector in India: An

A COMPARATIVE STUDY OF EFFICIENCY IN CENTRAL AND EASTERN EUROPEAN BANKING SYSTEMS

Economic Efficiency of Ring Seiners Operated off Munambam Coast of Kerala Using Data Envelopment Analysis

TESTING LENDING EFFICIENCY OF INDIAN BANKS THROUGH DEA

Benchmarking and Data Envelopment Analysis: An Approach to Rank the Best Performing Engineering Colleges Functioning in Tamil Nadu

Iranian Bank Branches Performance by Two Stage DEA Model

Measuring the Relative Efficiency of Banks: A Comparative Study on Different Ownership Modes in China

Efficiency Analysis Of Non-Life Insurance Companies In Terms Of Underwriting Process With Data Envelopment Analysis

EFFICIENCY EVALUATION OF BANKING SECTOR IN INDIA BASED ON DATA ENVELOPMENT ANALYSIS

Analysis of the Operating Efficiency of China s Securities Companies based on DEA Method

Estimating Technical Efficiency of Academic Departments of a Philippine Higher Education Institution

A Linear Programming Formulation of Macroeconomic Performance: The Case of Asia Pacific

Using Data Envelopment Analysis to Rate Pharmaceutical Companies; A case study of IRAN.

Zimbabwe commercials banks efficiency and productivity analysis through DEA Malmquist approach:

Data Envelopment Analysis in Finance and Energy New Approaches to Efficiency and their Numerical Tractability

Labour Cost Efficiency of Indian Banks: A Non-Parametric Analysis*

ASSESSING EDUCATION AND HEALTH EFFICIENCY IN OECD COUNTRIES USING ALTERNATIVE INPUT MEASURES. António Afonso and Miguel St.

Efficiency Measurement of Enterprises Using. the Financial Variables of Performance Assessment. and Data Envelopment Analysis

THE FINANCIAL PERFORMANCE AND CREDIT RISK OF MOLDOVAN AND PORTUGUESE COMPANIES USING DATA ENVELOPMENT ANALYSIS. Ana Paula Monte

PerformanceEvaluationofFacultiesataPrivateUniversityADataEnvelopmentAnalysisApproach

Comparison on Efficiency of Foreign and Domestic Banks Evidence from Algeria

CARDIFF BUSINESS SCHOOL WORKING PAPER SERIES

Efficiency of Macedonian Banks: A DEA Approach 5

INFORMS International Conference. How to Apply DEA to Real Problems: A Panel Discussion

International Journal of Management (IJM), ISSN (Print), ISSN (Online), Volume 4, Issue 1, January- February (2013)

MASTER OF SCIENCE IN MONETARY AND FINANCIAL ECONOMICS MASTERS FINAL WORK DISSERTATION ASSESSING PUBLIC SPENDING EFFICIENCY IN 20 OECD COUNTRIES

Measuring Technical Efficiency of the Iranian Post Company Using Data Envelopment Analysis (DEA)

On the Human Capital Factors to Evaluate the Efficiency of Tax Collection Using Data Envelopment Analysis Method

The Cost-Efficiency of French Banks

Antonella Basso - Stefania Funari

Non-parametric Approaches to Education and Health Expenditure Efficiency in the OECD 1

Data Envelopment Analysis (DEA) for evacuation planning

International Journal of Economics, Commerce and Management United Kingdom Vol. V, Issue 5, May 2017

Non-parametric Approaches to Education and Health Expenditure Efficiency in OECD Countries 1

Benchmarking Transnet Limited s Petroleum Pipelines

EFFICIENCY IN INTEGRATED BANKING MARKETS AUSTRALIA AND NEW ZEALAND

Operating Efficiency of the Federal Deposit Insurance Corporation Member Banks. Peter M. Ellis Utah State University. Abstract

Envelopment Methodology to Measure and Compare Subcontractor Productivity at the Firm Level

Efficiency of Swedish equity funds A DEA approach

Comparative study of Cost and Revenue efficiency in public sector banks in India DEA Approach

An Analysis of Revenue Maximising Efficiency of Public Sector Banks in the Post-Reforms Period

Measuring Efficiency of Australian Equity Managed Funds: Support for the Morningstar Star Rating

Assessing Public Spending Efficiency in 20 OECD Countries

A Stepwise-Projection Data Envelopment Analysis for Public Transport Operations in Japan. Peter Nijkamp b

The cost of mismanagement of gold production in Sudan

OPERATIONAL EXPANDITURE BENCHMARKING OF REGIONAL DISTRIBUTION UNITS AS A TOOL FOR EFFICIENCY EVALUATION AND DESIRED COST LEVEL ESTIMATION

Comparative Analysis of Technical Efficiency for Islamic versus Conventional Banks and its Determinants in Pakistan

The Relative Efficiency of Saudi Banks: Data Envelopment Analysis Models

Revenue Malmquist Productivity Index And Application In Bank Branch

Efficiency Evaluation of Thailand Gross Domestic Product Using DEA

EVALUATING THE EFFICIENCY OF EUROPEAN UNION INTEGRATION

Performance of Financial Expenditure in China's basic science and math education: Panel Data Analysis Based on CCR Model and BBC Model

A comparative Study on Intermediate Public Examination of Andhra Pradesh by Data Envelopment Analysis

Data Envelopment Analysis for Stocks Selection on Bursa Malaysia

A new inverse DEA method for merging banks

Application of Data Envelopment Analysis Method to Analyze and Improve Efficiency of Infrastructure Maintenance Strategies

A COMPARATIVE ANALYSIS OF ACCOUNTING AND FINANCIAL PRACTICES ASSOCIATED WITH EFFICIENCY OF COOPERATIVE RURAL BANKS IN SRI LANKA

Does Bank Performance Benefit from Non-traditional Activities? A Case of Non-interest Incomes in Taiwan Commercial Banks

Data Envelopment Analysis (DEA) Approach for the Jordanian Banking Sector's Performance

Efficiency and productivity change in the banking industry: Empirical evidence from New Zealand banks

Research Article A Two-Phase Data Envelopment Analysis Model for Portfolio Selection

Review of Middle East Economics and Finance

Bank Efficiency and Economic Freedom: Case of Jordanian Banking System

Efficiency of Islamic Banks in Sudan: A non-parametric Approach 1

Financial performance measurement with the use of financial ratios: case of Mongolian companies

Loughborough University Institutional Repository. This item was submitted to Loughborough University's Institutional Repository by the/an author.

Contents. Preface... Part I Single-Objective Optimization

Portfolio Selection using Data Envelopment Analysis (DEA): A Case of Select Indian Investment Companies

Share Performance and Profit Efficiency of Banks. in an Oligopolistic Market: Evidence from Singapore

Available online at ScienceDirect. Procedia Economics and Finance 12 ( 2014 )

Measuring Public Expenditure Efficiency. Yong Yoon Faculty of Economics Chulalongkorn University

New Quantitative Approaches to Asset Selection and Portfolio Construction. Irene Song

The impact of mergers and acquisitions on the efficiency of GCC banks

A DEA MEASURE FOR MUTUAL FUNDS PERFORMANCE

TECHNICAL EFFICIENCY OF EDUCATION SECTOR IN THE EU AND OECD COUNTRIES: THE CASE OF TERTIARY EDUCATION

PERFORMANCECONSISTENCY OF PRIVATE SECTORBANKS IN INDIA -A DEA APPROACH

Research Article An Empirical Study of Hybrid DEA and Grey System Theory on Analyzing Performance: A Case from Indian Mining Industry

THE PERFORMANCE OF AUCTION HOUSES

Technical efficiency and its determinants: an empirical study on banking sector of Oman

Mutual Fund Performance Evaluation using Data Envelopment Analysis with Higher Moments

Martingale Pricing Theory in Discrete-Time and Discrete-Space Models

EFFICIENCY OF PUBLIC EXPENDITURE ON EDUCATION: COMPARING CROATIA WITH OTHER NMS

Available online at ScienceDirect. Procedia Economics and Finance 33 ( 2015 )

International Journal of Academic Research ISSN: ; Vol.3, Issue-5(2), May, 2016 Impact Factor: 3.656;

An Examination of Technical, Pure Technical, and Scale Efficiencies in Indian Public Sector Banks using Data Envelopment Analysis

MULTIDIMENSIONAL PERFORMANCE OF LISTED COMPANIES AT THE PHILIPPINE STOCK EXCHANGE EMILYN C. CABANDA & FLORENCE P. BOGACIA

EFFICIENCY OF THE MACEDONIAN BANKING SECTOR

Decomposition of the Technical Efficiency: Pure Technical and Scale Efficiency of the Financial System

Earning Capacity, Efficiency and Poverty: A Study on Rural West Bengal and Orissa

Evaluating Total Factor Productivity Growth of Commercial Banks in Sri Lanka: An Application of Malmquist Index

The International Journal of Banking and Finance, 2007/08 Vol. 5. Number 2: 2008:

On the Distributional Assumptions in the StoNED model

Transcription:

Ranking Universities using Data Envelopment Analysis Bronwen Edge September 1, 2016 Bronwen Edge Data Envelopment Analysis September 1, 2016 1 / 21

Outline 1 Introduction What is DEA CCR Model BCC Model 2 The Data Overview 3 Scenario 4 Examples Adams CRS Adams VRS Adams Scottish Universities Choosing inputs and outputs 5 Discussion Advantages of DEA Limitations of DEA 6 Questions Bronwen Edge Data Envelopment Analysis September 1, 2016 2 / 21

Introduction Each year university ranking tables are released, A universities place in such tables greatly influences a student s decision. Can a personal ranking be made to suit a students individual needs? DEA a suitable method for this problem? Bronwen Edge Data Envelopment Analysis September 1, 2016 3 / 21

What is DEA Introduced by Charnes, Cooper and Rhodes in 1978. It is a benchmarking technique used to evaluate firms and assess their performance against their peers. DEA creates a frontier based on the most efficient firms which envelopes the inefficient firms. E and F are inefficient. Bronwen Edge Data Envelopment Analysis September 1, 2016 4 / 21

What is DEA Figure 1: To improve F efficiency a proportional decrease in input usage should be made to place F on the frontier while still producing the same outputs. E is the projected point for optimum efficiency for firm E. DEA involves constructing a hypothetical best case firm out of the existing firms, if any of firms are not as efficient as this best case one, they are inefficient. Collecting all the efficient cases together makes up the frontier, which then envelops the inefficient firms. Bronwen Edge Data Envelopment Analysis September 1, 2016 5 / 21

Notation DMU s Inputs X Outputs Q... DMU n x n q n Notation: There are m inputs, s outputs for each of the N firms, for the nth firm the inputs and outputs are represented by vectors x n and q n. Efficiency = Outputs Inputs Bronwen Edge Data Envelopment Analysis September 1, 2016 6 / 21

CCR Model Ratio Form Charnes, Cooper and Rhodes (1978) proposed a CCR model that had an input orientation and assumed constant returns to scale. Input orientation means that when trying to improve efficiency, inputs are reduced while outputs remain constant. As the returns are constant an increase/decrease in inputs will lead to an equivalent increase/decrease in outputs. Bronwen Edge Data Envelopment Analysis September 1, 2016 7 / 21

CCR Model Ratio Form We would like to measure the ratio of all outputs over all inputs. max u,v (u q n /v x n ), st u q j v x j 0, u, v 0, Set v x n = 1 to avoid an infinite amount of solutions. max u,v (u q n ), st v x n = 1 u q j v x j 0, u, v 0, j = 1, 2,..., N u =output weights v =input weights Bronwen Edge Data Envelopment Analysis September 1, 2016 8 / 21

CCR Model Envelopment Form Next we use the duality in linear programming, to derive the Envelopment form for this problem: min θ,λ θ, st q n + Qλ 0, θx n Xλ 0, λ 0. λ the constraint vector, θ n is the efficiency score of the nth firm. Involves fewer constraints. θ = 1 a DMU is efficient and is added to the frontier. Bronwen Edge Data Envelopment Analysis September 1, 2016 9 / 21

BCC Model The CCR model is used when an increase in input results in the same proportional increase in output. However most of the time this doesn t happen. In 1984, Banker, Charnes and Cooper suggested adjusting the CRS model to account for this, the Variable Returns to Scale (BCC) model. min θ,λ θ, st q i + Qλ 0, θx i Xλ 0, N1 λ = 1 λ 0. This model envelopes the data tighter and can tell you if a DMU is too large or small to be efficient. It is a way of only comparing DMU s of a similar size. Bronwen Edge Data Envelopment Analysis September 1, 2016 10 / 21

The Data Overview Choose to look at 69 university mathematics departments from across UK. Missing data. Different units. Massive variation in institutions. I sourced the data from: https://unistats.direct.gov.uk/ www.timeshighereducation.com Bronwen Edge Data Envelopment Analysis September 1, 2016 11 / 21

Scenario Adam is a Scottish mathematics student, he will be choosing his options for university soon. Adam is very academic and has been advised to apply for Oxbridge but doesn t really feel like they re for him. He is interested in exploring other options before he makes his decision. Adam doesn t want to accumulate a muscular amount of debt so hes looking to reduce his living costs. He s interested in value for money so studying at a Scottish university is attractive and high teaching standards is important. Adam is interested in working rather than further study. Adams hobbies include: Gunplay. Bronwen Edge Data Envelopment Analysis September 1, 2016 12 / 21

Examples 1 Inputs: Entry Requirements with teaching score Average annual fee Cost of private accommodation Outputs: Percentage of students with a good degree Average salary after 6 months Percentage of students working in a professional job after 6 months Bronwen Edge Data Envelopment Analysis September 1, 2016 13 / 21

Adams CRS Entry requirements with teaching score, average annual fee, cost of private accommodation against percentage of students with a good degree, average salary after 6 months and percentage of students working in a professional job after 6 months. Bronwen Edge Data Envelopment Analysis September 1, 2016 14 / 21

Adams VRS Entry requirements with teaching score, average annual fee, cost of private accommodation against percentage of students with a good degree, average salary after 6 months and percentage of students working in a professional job after 6 months. Bronwen Edge Data Envelopment Analysis September 1, 2016 15 / 21

Examples:Scottish Universities CRS Entry requirements with teaching score, average annual fee, cost of private accommodation against percentage of students with a good degree, average salary after 6 months and percentage of students working in a professional job after 6 months. Bronwen Edge Data Envelopment Analysis September 1, 2016 16 / 21

Adams Scottish Universities VRS Entry requirements with teaching score, average annual fee, cost of private accommodation against percentage of students with a good degree, average salary after 6 months and percentage of students working in a professional job after 6 months. Bronwen Edge Data Envelopment Analysis September 1, 2016 17 / 21

Choosing inputs and outputs Things to think about when selecting inputs and outputs: Fitting inputs and outputs to a students personal gives a better model. Try not to introduce bias, be objective! Try and avoid double counting, Avoid mutually exclusive outputs. Bronwen Edge Data Envelopment Analysis September 1, 2016 18 / 21

Advantages of DEA DEA can handle multiple input and output models, Peers are directly compared against peers, Inputs and outputs can have very different units. Works very well for objective inputs and outputs. Bronwen Edge Data Envelopment Analysis September 1, 2016 19 / 21

Limitations of DEA As DEA is a extreme point technique it is skewed by extreme values or outliers. Large problems can be computationally intensive. Choosing the right inputs/outputs/environmental factors is tricky when considering universities. Some may lead to double counting as they are not independent. Not convinced that DEA is suitable for this problem. Bronwen Edge Data Envelopment Analysis September 1, 2016 20 / 21

Questions Thanks for listening, Any Questions? Bronwen Edge Data Envelopment Analysis September 1, 2016 21 / 21