Casey Quinn
Centre for Health Economics
The University of York
York, YO10 5DD
phone: +44 01904 32 1411
fax: +44 01904 32 1402
email:
web:
Education
2003-2006The University of YorkYork
Doctor of Philosophy (Health Economics)
My doctorate focuses upon the use of copulas in health economics, introducing and explaining them with applications to specific statistical problems in health economics.
My supervisors are Prof. Andrew Jones (Dept. Economics) and Dr. Nigel Rice (Centre for Health Economics).
2001-2003The AustralianNationalUniversityCanberra
Master of Philosophy (Health Economics)
Two-year Master’ degree by research in health economics, specifically the private health insurance purchasing decision.
My supervisor was Prof. James RG Butler (National Centre for Epidemiology and Population Health).
1997-2000The University of SydneySydney
Bachelor of Commerce Hons (Econometrics)
Four-year Bachelor’s degree with majors in Economics and Econometrics, with Economics honours subjects and fourth-year Econometrics honours. My thesis was a theoretical examination of alternative specifications of joint distribution functions in Double Hurdle models.
My supervisor was Assoc. Prof. Murray D (Dept.Econometrics).
Professional memberships
- The American Statistical Association
- The Australian Health Economics Society, with whom I also held the office of Editor.
- The Econometric Society.
- The Health Economists’ Study Group
- The International Health Economics Association.
publications and other research
Quinn C (2005) Generalisable regression methods for cost-effectiveness using copulas, Working Paper 05/13 of the Health, Econometrics and Data Group at the University of York, York.
Banwell C Dance P Quinn C Davies R (2003) From the Ground Up a report on the need for an ACT-based rehabilitation, halfway house and counselling service run by the Construction, Forestry, Mining and Energy Union and the Building Trades Group, Canberra.
Butler J Quinn C (2003) Evaluation of the Commonwealth-funded Hepatitis C Education and Prevention Program, 1999-00 to 2002-03 a report prepared for the Commonwealth Department of Health and Ageing, Canberra.
Quinn (2003) Estimation of Private Health Insurance Purchasing, Master Thesis (unpublished), National Centre for Epidemiology and Population Health, The Australian National University, Canberra.
Quinn C (2003) The pasts and futures of private health insurance in Australia, Working Paper 47, National Centre for Epidemiology and Population Health, The Australian National University, Canberra.
Quinn C (2002) Private health insurance in Australia: where has it been, where might it have gone, and where might it go? Paper presented to the Conference of the Australian Health Economics Society, Sydney, 2002.
Quinn C (2000) An Examination of Copulae and Estimation of the Double-Hurdle Model, Honour Thesis (unpublished), Discipline of Econometrics and Business Statistics, The University of Sydney, Sydney.
Research Interests
- Copulas and measures of dependence as well as distribution theory more generally.
- Health economics; health insurance and health care financing.
- Health economics; lifestyles, mortality and self-assessed health.
- Cost-effectiveness analysis and economic evaluation.
- Panel data methods.
experience – teaching
2004/06The University of YorkYork
Teaching Assistant
Department of Economics and Related Studies
As a teaching fellow for the department my responsibilities principally involve the presentation and discussion of set exercises in seminars, which I grade as assessed work, as well as giving general guidance and consultation. I have taught courses in Statistical theory and introductory mathematics for first-year undergraduate economics students.
Health Economics for Health Care Professionals
As well as teaching in a classroom setting for the department, I have been engaged in the distance learning programme Health Economics for Health Care Professionals by Distance Learning.
This involved providing online discussion through a chat room/bulletin board system, requiring a unique form of interaction; tutoring and guiding mature-aged health care professionals with little to no background in economics. I also supervise off-campus exams for the programme.
Department of Mathematics
The department of Mathematics has also engaged me as a marker on first-year statistics exercises and examinations.
2002 The AustralianNationalUniversityCanberra
Teaching Assistant
School of Economics
Faculty of Economics and Business
I was the teaching assistant for my supervisor’s taught class in the economics department. This was a course in health economics, principally from a microeconomic perspective with standard concepts built into the framework of the Australian health care system. It was for 3rd-year and Master class economics students.
Typically this involved running tutorials. I was also responsible for revision classes and exam supervision. While my supervisor was absent for his own research I also gave the lectures.
2001/02TheAustralianNationalUniversityCanberra
Teaching Assistant
National Centre for Epidemiology and Population Health
I was substitute-lecturer for a post-graduate course on the Principles of Health Economics, in the absence of the course lecturer. These were two-hour lectures to a small class of Masters and Doctoral students, also made up mostly of health care professionals with limited economic backgrounds. I also supervised examinations in the lecturer’s second absence.
2000The University of SydneySydney
Teaching Assistant
Discipline of Econometrics and Business Statistics
School of Economics and Political Science
Faculty of Economics and Business
As a teaching assistant for first-year econometrics I prepared and ran workshops; part lecture and part tutorial according to the system in the department, for classes of up to 90 students. As the introductory course, running the entire first year, this included all basic concepts in econometrics, including some regression. I also taught the same course with a semester lag, for international exchange students, and repeating students.
I was also responsible for administration of exams, as well as grading assignments and examinations.
experience – research and other
2003/4The AustralianNationalUniversityCanberra
Consultant
National Centre for Epidemiology and Population Health
I have been engaged twice, along with my supervisor Prof. Jim Butler, by the Australian Department of Health and Ageing to provide programme evaluation. These were programmes for the prevention of hepatitis C, and falls in the elderly. In each instance we were contracted by the Department in order to fulfill funding review requirements of the programmes.
In a more clinical context, I have written on the methodology of cost-effectiveness analysis and economic evaluation in clinical trials (see publications), showing an augmented method for cost-effectiveness analysis with more accurate estimation. This is also ongoing work with another member of the Centre for Health Economics.
2003-The University of YorkYork
Web Administrator
Department of Economics and Related Studies/Centre for Health Economics
Health, Econometrics and Data Analysis Group
Since joining the Centre for Health Economics, I have been responsible for the then York Seminars in Health Econometrics website. I have since assisted Nigel Rice and Andrew Jones in formalising the seminar series into a dedicated, funded research group ( and establishing its presence online, as well as contributing to promotional material and the associated working paper series.
As well as being responsible for the construction and content of the HEDG website, I maintain the broader site of the Health Economics Resource Centre.
2003-The University of YorkYork
Health, Econometrics and Data Analysis Group, and
2002-2005Australian Health Economics Society
I was, from 2002-2005, Editor for the Australian Health Economics Society. This office entailed the collection and editing of paper presented at the annual conference of the society for publication.
As part of my duties in the Health, Econometrics and Data Group I also edit, to a lesser degree, papers submitted to the HEDG working paper series. This involves only straightforward proofing and formatting.
2003The AustralianNationalUniversityCanberra
Research Assistant
Centre for Aboriginal Economic Policy Research
While at the ANU I assisted Dr. Boyd Hunter with analysis of court appearances data, to examine the relationship with criminal behaviour of Indigenous people.
2001-2003The AustralianNationalUniversityCanberra
Research Assistant
National Centre for Epidemiology and Population Health
My research in this capacity was varied, combining both research and annotation of literature, as well as data analysis. Topically I have –
- Aided a review of a report for the Northern Ireland Research and Development Office on Health and Personal Social Services in Northern Ireland.
- Provided statistical analysis for a report on gambling, alcohol and other drug dependency among workers in the building, forestry and construction industries in the ACT (ongoing, and I am co-author also of the completed report).
- Provided preparation for Genetics and Financial Services: Threat or Opportunity? Conference sponsored by the Centre for Actuarial Research, AustralianNationalUniversity, Canberra 2002.
Data Analysis Officer
National Service Development
During this time I was the resident statistician for CRS Australia. In this capacity I was responsible for the purchase and collation of data relevant to the rehabilitation and employment industry from various statistical bureaus and government offices.
My principle responsibility was the analysis and forecasting of potential service needs and subsequent funding distributions within the organisation. I contributed to a majority of the organisation’s 2001-02 budget and business plan by creating their funding model.
While there I was responsible for the creation and maintenance of several internal databases which were used for business planning, as well as the creation of an ongoing compendium of states and trends within the organisation over the previous decade.
rEFERENCES
Prof. Andrew Jones
Department of Economics and Related Studies
The University of York
Email:
Web:
Dr. Nigel Rice
Centre for Health Economics
The University of York
Email:
Web:
Assoc. Prof. Murray Smith
Discipline of Econometrics and Business Statistics
School of Economics and Politics Science
Faculty of Economics and Business
The University of Sydney
Email:
Web:
Abstracts
Generalisable regression methods for cost-effectiveness using copulas
HEDG Working Paper 05/13
Objectives: Covariate explanation of clinical trial cost and outcome is critical to allow reliable estimates of cost-effectiveness. The ordinary simultaneous equations approach however must specify a bivariate distribution for both cost and health outcome that is not typically a product of the best-fitting marginal distribution of each. This paper advocates estimating costs and outcomes simultaneously using copulas to model conditional dependence. The copula is a function that maps univariate marginal distributions of any to some joint distribution.
Methods: Copulas are used to fit the bivariate distribution of the simultaneous model for individual cost and outcome in a clinical trial for hysterectomy. These are used to generate counter-factual outcomes and individual-level incremental net benefits due to each procedure, as well as replicating non-parametric techniques for comparison with standard methods.
Results: Parametric results from the use of copulas compared to an ordinary Seemingly Unrelated Regression model show better fit with consistent estimates, allowing for the fact that the data is drawn from an underpowered clinical trial. Results also show that estimated coefficients vary in size, sign and statistical significance in different arms of the clinical trial. Non-parametric results compared with standard cost-effectiveness techniques also show more precise estimates of incremental net benefit.
Conclusions: Regression-based approaches to cost-effectiveness have the potential to overcome a lot of the limiting assumptions made using non-parametric approaches. By using known information on covariates we can get more precise estimates of the parameters used in standard cost-effectiveness analysis, more precise posterior information and more precise posterior probabilities. Using copulas generates more precise estimation of conditionally-dependent marginal effects.
Keywords: Cost-effectiveness analysis, incremental net benefit, simultaneous equations, copula.
JEL classification: C1, C31, I10
Copulas for Health Economics: An introduction, explanation and some applications
HEDG Working Paper 06/02
This is intended as a brief introduction, almost a user's guide, to copulas and multivariate dependence issues within a health economics context. A copula is best described, as in Joe (1997), as a multivariate distribution function that separates each marginal distribution both from every other marginal distribution, and from the dependence between their associated random variables.
The research presented here will make its own contributions to the development of copulas as a methodology, but more importantly will make deliberate inroads into health economic applications of copulas. To do this, we will consider common analytic problems faced by health economists.
The differences between the copula methodology and existing alternatives will be discussed, and a generalisable, systematic approach to estimation will be provided. It is hoped that in this way copulas will be not only introduced to health economists, but introduced in a statistically relevant manner so that they can be applied easily and with less study of what has been a long process of technical development.
Keywords: copulas, simultaneous equations, multivariate distributions, dependence, simulation.
JEL classification: C1, C3, I1
Alternative methods for the efficient estimation of systems of health equations
There are instances in health economics in which multivariate models are required for the analysis of several outcomes. Some of these are recognisable health-economic analogs of economic problems: stochastic frontier analysis for productivity and production cost, sample selection and so forth.
This paper presents alternative models for a system of equations representing mortality, health and lifestyles, where they are interdependent. We use a standard system of probit models, as well as two multivariate copulas: a multivariate FGM copula with general dependence and a multivariate Archimedean copula with some negatively-correlated bivariate margins, achieved via mixtures of Laplace transforms. Finally we use the method of Inference Functions for Margins to compare with maximum likelihood, to determine the importance of having a closed-form multivariate distribution in practice.
Keywords: health, lifestyle, mortality multivariate models, copulas, inference functions.
JEL classification: C1, C3, I1