CURRICULUM VITAE - Apostolos-Paul N. REFENES
Apostolos-Paul
Refenes (Bsc Mathematics & Computing 1984, PhD Computing 1987) is Professor
of Financial Engineering and Director of the Financial Engineering Research
Centre at Athens University of Business & Economics. He has held previous
academic appointments at London Business School (Associate Professor),
University College London (Senior Research Fellow), and the University of
Athens (Visiting Professor), and various professional appointments including OPAP
International Ltd (Chief Executive), OPAP S. A. (Board member, Non-executive
Director), Hellenic Competition Commission (Member), UK Cabinet Office (Member
Technology Foresight Panel on financial services), and the DTI (Scientific
Advisor). He has consulted for many financial institutions including
Morgan-Stanley, CitiBank, Barclays, Dresdner, BNP, Societe General, Smith New
Court, Golden Cross, KGAL, Bank of Greece, OTEestate, etc. and other
organisations including the European Commission.
Author of over
100 papers and editor of six books on the subjects of neural computing and
financial engineering applications. Recipient of research awards in excess of $15m from several public and private funds. Associate editor
of the Intelligent Systems in Accounting and Finance Journal, International
Journal of Computational Intelligence and Organisations, guest editor of
the Journal of Forecasting, series editor Studies in Computational
Finance, and a serving member of the editorial board of the Neural
Computing & Applications Journal. Programme committee member in several
International conferences including World Congress on Neural Networks, International
Conference on Artificial Neural Networks, International Conference on
Neural Information Processing Systems, Euromicro, the IEE annual
conference on ANNS, and ICANN. Founded the international conference
on Neural Networks in the Capital Markets (NnCM) and served as general
chair for NnCM-93 and NnCM-95, Computational Finance 1997,
International Chair for the Joint IEEE/IAFE conference on Computational
Intelligence in Financial Engineering (CIFEr). Invited speaker at many
international symposia.
Teaching
interests include non-parametric statistics, neural networks, financial
econometrics, financial mathematics, and computational finance. His classes
include MBA, Masters in Finance and PhD students at London Business School as
well as MSc and BSc in computer Science at Athens University of Economics &
Business and University College London. Short executive courses in financial
econometrics, advanced quantitative methods, advanced data analysis and
forecasting have been developed over the years. Invited to give university
lectures and short executive courses in over 20 countries in Europe, Asia,
Australia, North and South America.
Current
research on Neural Networks, Financial Engineering and Computational Finance is
supported by the ESRC, the DTI, ESPRIT, VALUE, and privately by several
companies in the finance sector. Work on neural network design methodology,
model identification, and estimation procedures is cited regularly for a number
of years. Applied work has also included tactical asset allocation, factor
models for equity investment, dynamic risk management, nonlinear cointegration,
exchange risk management, etc. Research papers have appeared in such journals
as Journal of Forecasting, Journal of Futures Markets, Journal of
Risk Finance, IEEE Trans on Neural Networks, Neural Networks,
Neural Computing and Applications, Neurocomputing, Risk, Defense
Economics, etc. Topical work has been reported in Scientific American,
The New Scientist, Nature, Risk, IEEE spectrum, and
the press The Financial Times, The Times, The Independent,
The Guardian, The Daily Telegraph, and others. Listed in “who is who in the world”.
PERSONAL
DETAILS
|
|
|
|
|
|
NAME |
Apostolos-Paul N.
Refenes |
|
|
ADDRESS |
40, Menelaou Street, Voula, Athens |
|
|
TELEPHONE |
(++30) 210 82 03 660 |
|
|
MARITAL STATUS |
Married |
|
|
AGE |
48 |
CURRENT
POSITION
Professor
of Financial Engineering, Athens University of Economics & Business (AUEB)
Director,
Financial Engineering Research Unit, AUEB.
EDUCATION
PhD in
Computer Science (1987).
BSc (Hons)
in Mathematics and Computing (1984).
POSITIONS
HELD:
RESEARCH
|
(1995 - 2000) |
Associate Professor in Decision
Science, London Business School |
|
(1996 - 2000) |
Director, Computational Finance
Programme, London Business School |
|
(1994 - 1996) |
Director, NeuroForecasting Club,
London Business School |
|
(1989 - 1993) |
Senior Research Fellow, University
College London |
|
(1987 - 1989) |
Research Associate, University
College London |
|
|
|
|
(1994 - 1996) |
Visiting Professor, University of
Athens |
|
(1994 - 1995) |
Visiting Senior Research Fellow,
London Business School |
|
(1990 - 1992) |
Visiting Science Advisor, Department
of Trade & Industry |
ADMINISTRATIVE
|
(2003 - 2005) |
Chair, Student Club, AUEB |
|
(2001 - 2003) |
Deputy Chair, Department of
Management Science, AUEB |
|
(1990 - 1991) |
Chair, CEC Advanced Informatics in Medicine Working Group |
|
(1990 - 1991) |
Chair, DTI Mission to Assess Japanese FGCS Programme |
|
(1989 - 1990) |
Member, ESPRIT experts group for the
Parallel Computing Action |
PROFESSIONAL
|
(2004 - ) |
Chief Executive, OPAP International
Ltd. |
|
(2004 - ) |
Board Member, Hellenic Competition
Commission |
|
(2004 - 2004) |
Board Member, OPAP S.A.,
Non-Executive Director, |
|
(1993 - 1999) |
Chairman, Hughes Financial Analytics
Ltd |
|
(1996 - 2000) |
Panel Member, Cabinet Office(UK),
OST; Financial Services ForeSight |
EDITORIAL
|
Assoc. Editor |
|
Intelligent Systems in Accounting & Finance, Wileys,
(1998 - 2000) |
|
Assoc. Editor |
|
Int. Journal of Computational Intelligence &
Organisations, IJCIO (95- 00) |
|
Guest Editor |
|
Journal of Forecasting, Special Issue,
co-edited H.White, Vol. 17(1998) |
|
Editorial Board |
|
Neural Computing & Applications Journal,
Springer Verlag, (1991 - ) |
|
Editorial Board |
|
Knowledge Based Intelligent Engineering Systems (2001
- ) |
|
Editor |
|
Neural Networks in the Capital Markets, Wiley & Sons, Book
(1995) |
|
Editor |
|
Proc. First Int. Wrksp. "Neural
Networks in the Capital Markets", (1993) |
|
Series Editor |
|
Computational Finance, Kluwer Academic, Book Series
(1998 -) |
|
Co-editor |
|
Decision Technologies for Financial Engineering, WSP, Book (1997) |
|
Co-editor |
|
Neural Networks in Financial Markets, Proc. NnCM96 WSP, Book (1996) |
|
Co-editor |
|
Decision Technologies in Computational Finance, Kluwer,
Academic Proc. Computational Finance 1997, Book (1998). |
CONFERENCE
ORGANISATION
|
General Chair |
COMPUTATIONAL
FINANCE 1997, London (October 1997). |
|
General Chair |
3rd International Conference on
"Neural Networks in the Capital Markets", NnCM, London (Oct. 1995). |
|
General Chair |
International Workshop on
"Neural Networks in the Capital Markets" London Nov. 18-19 (1993). |
|
International Chair |
Joint IEEE/IAFE Int. Conf. on "Computational
Intelligence in Financial Engineering", New York, Spring 1995. |
|
Session Chair |
"Dynamical Systems in Financial
Engineering", WCNN (1995):- World Congress on
Neural Networks Washington DC (1995). |
|
Session Chair |
Sixth European Congress on
Intelligent Techniques & Soft Computing, Non-Parametric Methods in
Financial Econometrics, Aachen, Sept. 98. |
|
Programme Committee |
Computational Finance 1998, New York,
NYU, Stern (1994 - ) |
|
Programme Committee |
International ICSC Symposium on Soft
Computing in the Financial Markets), June 1999, Rochester, NY, USA, (1999). |
|
Programme
Committee |
Engineering
Applications of Neural Networks (EANN'99),
13-15 September 1999, Warsaw, Poland. |
|
Programme
Committee |
World Congress on Neural Networks, WCNN (1995 -) |
|
Programme
Committee |
International Conference on Artificial Neural Networks,
ICANN (1995 -) |
|
Programme
Committee |
International Conference on Neural Information Processing
Systems, ICONIPS (1995 -) |
|
Programme
Committee |
IEE International Conference on ANNS, IEE (1992 - 1996) |
INVITED/PLENARY
TALKS - Conferences
|
Keynote
address |
ICANN' 93,
International Conference on Neural Networks, "Neural Networks in the
Capital Markets", Amsterdam, (Sept. 1993). |
|
Keynote address |
NIPS'
93,
"Non-linear methods in Financial Engineering", Denver, Colorado, (Dec. 1993). |
|
Keynote
address |
IEEE,
Int. Conf. Computational Intelligence, Perth, Australia (Nov. 95) |
|
Invited
Speaker |
IIR
Quantitative Portfolio Investment Techniques, London, (Oct
1999) |
|
Invited
speaker |
Knowledge
Based Intelligent Information Systems Engineering, Osaka, Japan, (Sept.
2001). |
|
Invited
speaker |
EQMC,
Non-parametric Methods in Quantitative Marketing, Madrid, July 1998. |
|
Invited speaker |
NnCM 96, (Neural Networks in the
Capital Markets). Pasadena, CA, (Nov. 1996) |
|
Invited speaker |
ICONIPS96, (Int.
Conf. on Neural Information Processing). Hong-Kong, (Sept. 1996). |
|
Invited speaker |
CIFEr’95 (Computational Intelligence in
Financial Engineering). The first joint IEEE/IAFE int. conference on the
topic. NY, NY, (April 1995). |
|
Invited speaker |
RISK, Risk Conference on Model
Risk, "Evaluating and Managing Model Risk in the Non linear
Context", NY, NY (Oct. 1995). |
|
Invited speaker |
WCNN’94,
World Congress on Neural Networks, "Neural Networks in Investment
Management", San Diego June 1994. |
|
Invited speaker |
IBC, Fifth Annual Forum,
Advanced Technologies for Trading & Asset Management, "Nonlinear
Data Analysis and Forecasting in Investment Management:, New York July 20, 21
(1994). |
|
Invited speaker |
IBC, 5th
Annual Symp., Intelligent Systems in Business & Finance, "Neural Networks
in Financial Engineering", London Feb. (1994). |
|
Invited speaker |
IIR'
93,
Institute for International Research, conference on "Software tools fur
portfolio management und trading", Frankfurt, (April. 1993). |
|
Invited speaker |
IIR'
93,
Institute for International Research, conference on "Modernes Portfolio Management", Frankfurt, (Sept.
1993). |
|
Invited discussant |
Economic Notes (Risks Involving Derivatives), Sienna, (Dec. 1996). |
Invited speaker to many other
conferences and workshops in the UK and abroad including Global Derivatives 95
(Paris), RISK (NY), BNCNN-95 (Curitiba, Brasil), NIPT-91 (Tokyo), IWIC (USSR),
BCS, International Neural Networks Society, NCAF, IBC, Cambridge University
"Advances in Options Research", etc).
AWARDS & DISTINCTIONS
|
Best
Paper |
INQUIRE
– Institute for Quantitative Investment Research (1996) |
|
Best
Paper |
ICONIP - Int. Conf. On Neural Networks (1995) |
|
Best
Student |
University of North
London - Best Student Award, Department of Mathematics & Computer
Science (1982) |
REFEREEING
ACTIVITIES
¨ UK
Research Councils ESRC/EPSRC
¨ Commission
of the European Communities ESPRIT
¨ Hong Kong
University Grants Committee
¨ Cyprus
University Grants Committee
¨ IEEE
Trans. on Neural Networks
¨ IEEE
Trans. on Knowledge and Data Engineering
¨ Neural
Networks
¨ Neural
Computation
¨ Neural
Computing and Applications
¨ Neurocomputing
¨ Neural
Information Processing Systems
¨ Pattern
Analysis and Applications
¨ Computational
Statistics & Data Analysis
¨ Management
Science
¨ International
Journal of Forecasting
¨ European
Journal of Operational Research
¨ Computers
and Operations Research
¨ Journal of
Business Finance and Accounting
¨ Journal of
Defence Economics
¨ Journal of
Mathematics Applied to Business and Industry
RECENT
MEDIA
Topical work has been reported in professional media
¨ Scientific
American
¨ The New
Scientist
¨ Nature
¨ Risk
¨ IEEE
spectrum
¨ Canadian
Business Magazine
¨ Journal of
Global Investment
¨ Managed
Derivatives
¨ Listed in
who-is-who in the World
and the national press
¨ The
Financial Times
¨ The
Independent
¨ The Guardian
¨ The Daily
Telegraph
¨ Machine
Intelligence News
¨ Expert
System Applications and others.
CONSULTING
¨
CITIBANK
¨ Morgan
Stanley
¨ Barclays
BZW
¨ Credit
Lyonaisse
¨ Societe
Generale
¨ Dresdner
Bank
¨ Deutsche
Morgan Grenfell
¨ Reuters
Plc
¨ County
NatWest Investment Management
¨ Smith New
Court
¨ Golden
Cross
¨ Bradford
& Bigley Building Soc.
¨ Abbey
National
¨ Barclays
UKBS
¨
ECONOSTAT
¨ Thinking
Machines
¨ Shell
¨
IBM
¨ UK Department of trade & Industry
¨ EU ESPRIT advisory board
TEACHING
Executive
Courses
¨ Mastering
Advanced Quantitative Methods, Athens University of Economics and Business
¨ Financial
Engineering and Risk Management, GC, Brazil
¨ Pricing
Options, Futures and other Derivative Securities with Nonparametric Methods,
Golden Cross
¨ Factor Models
for Tactical Asset Allocation, Citibank, Singapore
¨ Advanced
Quantitative Investment Methods, Forum, Frankfurt.
¨ Advanced
Forecasting Methods for Financial Engineering, London Business School
¨ Neural
Networks in Financial Economics, Int. Center for Monetary & Banking
Studies, Geneva
¨ Tactical Asset
Allocation, International Faculty of Finance, London
Graduate
Courses
¨ Financial
Mathematics (Stochastic Processes, stochastic flows and differential equations)
¨ Financial
Econometrics (Time Series, ARCH/GARCH, State-Space
Models, Neural Networks)
¨ Stochastic
Optimisation and Genetic Algorithms
¨ Computational
Finance (Numerical methods, Re-sampling, Monte-Carlo, Bootstrap Statistics)
¨ Uncertainty
Analysis and Hypothesis Testing
Undergraduate
Courses
¨ Foundations of
Investment Management, HK
¨ International
Investment Decisions in Emerging Markets, Singapore
¨ Systems
Analysis, UCL
¨ Networks and
Architectures, UCL
¨ Neural
Networks, UCL
DOCTORAL
DISSERTATIONS
¨ Dotsis G., “Jumps
and Estimation Risk in Finance and Decision Making”, Athens University of
Economics & Business, PhD, (2006).
¨ Psychoyios D. “Volatility
Risk Hedging”, Athens University of Economics & Business, PhD, (2006).
¨ Skintzi V., “Dynamic
Correlation Models”, Athens University of Economics & Business, PhD,
(2004).
¨ Towers N.
“Evolutionary Methods for Decision and Risk Analysis in Active Investment
management”, London Business School, PhD (2000).
¨ Bolland P., “Robust
Neural Estimation and Diagnostics”, PhD, London Business School, (June 1998).
¨ Bentz Y.,
“Identifying and Modelling Conditional Factor Sensitivities: Applications in
Equity Investment”, PhD, London Business School, (Nov. 1999).
¨ Burgess A. N., A
Computational Intelligence Methodology for forecasting noisy, non-stationary
time-series, London Business School, PhD, (Nov. 1999).
¨ Holt W., “Statistical
Diagnostics and Test Procedures for Neural Models”, London Business School,
PhD (Feb. 1999).
¨ Pandelidaki S., “Neural
and Econometric Models for Sales Forecasting”, London Business School, PhD
(Nov. 1998).
¨ Zapranis A., “A
Methodology for Neural Model Identification, Variable Selection, and Adequacy
Testing”, PhD, London Business School, (June 1997).
¨ Azema-Barac M., “Parallel
Neural Network Architectures”, PhD University College London, (1994).
¨ Balou A., “A Basic
Object Oriented Platform for the execution of high-level OO languages”, PhD
University College London, (1995).
¨ Oliveira C., “A
Distributed Object-Oriented Machine for Parallel Processing”, PhD University
College London (1994).
RESEARCH
Since 1984 Refenes has been working on
dynamical systems theory and developed neural network applications in image
understanding, voice recognition, medical diagnosis, and database marketing.
Current research on methodology at the Decision Technology Centre, London
Business School deals with the development of nonlinear methods for data
analysis and forecasting. The main research themes cover the following :
Nonparametric
models & machine learning: non-parametric model estimation and
learning procedures based on neural networks.
Model
selection / specification: identification procedures for
mispecified (neural network) models; and diagnostics/residual analysis for
(non-linear) model (mis-) specification.
Hypothesis
testing & confidence intervals: development of distribution theory
for hypothesis testing and confidence intervals on parameter/variable
significance estimation.
Robust
model estimation: outlier- and leverage-resistent estimation procedures for
neural models and diagnostics for outlier/leverage identification in the
context of nonlinear models.
Parameter
sensitivity & prediction uncertainty: model sensitivity to
sampling variance and parameter perturbations. Bounds for prediction
uncertainty.
Nonlinear
cointegration: development and identification of nonlinear models with
error correcting terms on cross-sectional as well as time series data.
Generalised
Nonlinear Least Squares Models: development and application of GLS
methods for nonlinear models to deal with problems of stationarity, level
changes, etc.
For the past five years he is working
on non-linear methods for data analysis and forecasting in the context of
financial engineering, and business applications. He has been awarded external
research grants of over $10m and led teams in several research projects
including machine learning, non-linear systems for currency trading, tactical
asset allocation, and portfolio management. In the past three years as part of
the NeuroForecasting Research Programme at London Business School he directed
research and led teams working on the following projects:
Factor
Models for Tactical Asset Allocation: Factor
models are widely used in portfolio management. This project extended the
approach to tactical asset allocation whereby performance differentials between
the main asset classes (bonds vs
equities vs cash) can be explained in terms of changes in fundamental economic
and financial variables. This approach relaxes the assumptions on linearity and
uses neural networks instead of regression analysis to model relative
performance between the main asset classes on the basis of their exposure to a
set of (17) economic and financial factors. With: Postel (Hermes) Investment Management.
Arbitrage
Models for Tactical Asset Allocation: Statistical
arbitrage models are finding increasing use in tactical asset allocation as an
alternative to factor models. The basic idea is to exploit short-term pricing
anomalies between different asset classes. A model for the UK, exploiting daily
pricing anomalies between equities and gilts was completed on January 1995. With: Societe Generale.
Factor
Models for Equity Investment:
Linear
factor models are widely used in equity investment. By relaxing the assumptions
on linearity we use neural networks to model stock returns on the basis of
stock exposure to fundamental factors, financial ratios and cyclicity
indicators. The models are applied to a universe of stocks drawn for the
CAC-40. These non-linear factor models are then used to construct portfolios
which are immune and/or sensitive to given factor exposures by choosing the
weights so that the partial derivatives of the portfolio return over the chosen
factor exposures are set to the desired values. With: Societe Generale and Banque Nationale de Paris.
Nonlinear
Cointegration in European Equity Futures:
A
nonlinear co-integration model of the FTSE with a basket of European indices
(including DAX, EoE, CAC, and SMI) was developed on daily data. The residuals
of the cointegration are modelled as a nonlinear function of exogenous
variables (e.g. interest rate volatility, oil price changes, etc) selected via
ANOVA and neural network analysis. With: CitiBank.
Forecasting
Intra-day Volatility for Option Pricing:
Multivariate
neural models are developed to produce estimates of implied volatility to be
utilised in the context of option pricing for futures contracts. High frequency
tick-data from the Ibex-35 is used to develop the methodology. The neural
networks give incremental value in terms of forecasting accuracy over
time-series models and regression analysis. Sensitivity analysis is used to
verify the plausibility of the neural models. With: CitiBank.
Modelling
Quarterly Returns on the FTSE-ALSH and S&P 500: Neural
networks are utilised to model quarterly returns on the FTSE-ALSH and S&P
500 on the basis of fundamental factor changes (e.g. dividend yield, business
cycle, etc.). Up to eight fundamental variables are selected from a universe of
20 candidate variables, using regression and neural network analysis to
construct parsimonious models. Project focuses on outlier and
leverage-resistant neural network modelling methodology. With: Henderson Administration.
Term-structure
Models of Eurodollar Futures: Neural networks are used to
model the "volatility factor" in the term-structure of Eurodollar
futures. The "volatility factor" is the third principle component
which represents a flexing of the yield curve on a portfolio of short, medium
and long maturity contracts which has been immunised against parallel shifts
and rotations. This component is shown to be mean-reverting and it is linked to
volatility amongst other factors. The neural network model estimates variations
in this component which are then used as signals to reset the portfolio. With: CitiBank.
In his former position as senior
Research Fellow at University College London he led teams in many projects.
PUBLICATIONS
books
and special issues
[1]
Refenes A-P. N., and White H. (ed), "Neural Networks
and Financial Economics", Journal of Forecasting, special issue,
Vol. 17, 5-6., (1998).
[2]
Refenes A-P. N., Burgess A. N. and Moody J., (1998)
“Decision Technologies for Computational Finance”, Proc. Computational Finance
1997, Kluwer Academic, ISBN Hardback: 0 7923 8308 7; ISBN Paperback 0 7923 8309
5
[3]
Refenes A-P. N., Abu-Mostafa Y., Moody J., and Weigend A.
(ed), "Neural Networks in Financial Engineering", World Scientific,
Singapore, (1996), ISBN 981-02-2480x.
[4]
Refenes A.-P N. (ed), "Neural Networks in the Capital
Markets", Wiley & Sons, Chichester, (1995), ISBN 0-471-94364-9.
[5]
Weigend A, Abu-Mostafa Y. and Refenes A-P. N., (ed),
"Decision Technologies for Financial Engineering", World Scientific,
Singapore, (1997), ISBN 981-02-3123-7.
[6]
Zapranis A. D. and Refenes A-P. N., “Principles of Model
Identification, Selection and Adequacy: with Applications in Financial
Econometrics”, (1999), Springer-Verlag, ISBN1-85233-139-9.
[7]
Refenes A.-P. N. (ed) “Quantitative Methods in Finance”,
Typothito-George Dardanos, ISBN 960-402-173-7, Athens (2004)
journals
[8]
Skintzi V. D. and Refenes A-P. N. “Implied Correlation
Index: A new measure of Diversification”, Journal of Futures Markets,
Submitted August 2003, Accepted March (2004).
[9]
Skintzi V. D., Skiadopoulos G, and Refenes A-P. N. “The
effect of misestimating correlation on Value-At-Risk”, The Journal of Risk
Finance, 73(1), Submitted July 2003, Accepted February (2004).
[10] Skintzi V.
D. and Refenes A-P. N., "Volatility spillovers and dynamic correlation in
European Bond Markets”, Journal of International Financial Markets,
Institutions & Money, (submitted, accepted, 2004).
[11] Carapeto
M., Holt W., and Refenes A-P. N. “On model complexity and selection”, Journal
of Statistical Computation and Simulation, 73(1), pp. 45-47, (2003).
[12] Refenes
A-P. N. and Holt W. “Forecasting Volatility with Neural Regression: a
contribution to model adequacy”, IEEE Trans. On Neural Networks, Vol 12,
No.4, 850-865, (July 2001).
[13] Refenes
A-P. N., and Zapranis A. D. "Neural Model Identification, Variable
Selection and Model Adequacy", Journal of Forecasting, Vol. 18,
299-332, (1999).
[14] Refenes
A-P. N., Burgess A. N., and Bentz Y., "Neural Networks in Financial
Engineering: a study in Methodology", IEEE Trans on Neural Networks,
Vol. 8, No. 6, pp. 1222-1267, November 1997.
[15] Refenes
A-P. N., Bentz Y. Bunn W. D., Burgess A. N. and Zapranis A. D,
"Backpropagation with Discounted Least Squares and its Application to
Financial Time Series Modelling", Neurocomputing, Vol. 14, no. 2,
pp. 123-138 (Feb. 1997).
[16] Refenes
A-P. N., Gonzales Miranda F. and Burgess A. N., "Intraday Volatility
Forecasting Using Neural Networks. A Comparative Study with Regression Models",
IJCIO (accepted 1996, to appear 1996) Vol. 1:2, pp. 1-56.
[17] Kolias C.
and Refenes A-P. N. “Modelling the Effects of Defence Spending Reductions Using
Neural Networks: Evidence from Greece”, Journal of Peace Economics
and Public Policy, vol. 3. no. 2, pp. 1-12, (1996).
[18] Refenes
A-P. N., 'Neural Networks in Investment Management: Testing Strategies &
Performance Metrics', Neural Computing & Applications (revised Sept.
1994, accepted May 1995).
[19] Burgess A.
N., and Refenes A-P. N. “Modelling Nonlinear Moving Average Processes using
Neural Networks with Error Feedback: An application to implied volatility
Forecasting”, European Journal of Signal Processing, Vol. 74 , Issue (1998).
[20] Refenes
A-P. N., Kollias C., and Zapranis A. N., "External Security Determinants
of Greek Military Expenditure: An Empirical Investigation Using Neural
Networks", Journal of Defence Economics, vol. 6. pp. 27-41 (1995).
[21] Refenes
A-P. N., Francis G., and Zapranis A. D., "Stock Performance Modeling Using
Neural Networks: A Comparative Study with Regression Models", Neural
Networks Vol. 7, No. 2, pp 375-388 (1994).
[22] Refenes
A-P. N., "Neural Networks: forecasting Breakthrough or just a passing
fad", International Journal of Forecasting", 10(1994) 43-46.
[23] Refenes
A-P. N., and Azema-Barac M., "Neural Network Applications in Financial
Asset Management", Neural Computing & Applications, Vol. 2.,
no. 1, pp. 13-39. (1994).
[24] Refenes
A-P. N., et al "Currency
Exchange rate prediction and Neural Network Design Strategies", Neural
computing & Applications, Vol 1, no. 1., (1993).
[25] Tuv E.,
& Refenes A-P. N., "Removal of Catastrophic Noise in
Hetero-associative Training Samples", Microprocessing and
Microprogramming vol 38., pp. 697-704. (1993).
[26] Refenes
A-P. N., "N-Expression Implementations for Integrated Symbolic and Numeric
Processing", North-Holland Future Generation Computer Systems vol.
3, no. 3, pp.161 - 187, (Sept. 1987).
[27] Refenes
A-P. N, "Message-passing via Singly Buffered-channels: an Efficient and
Flexible Communications Control mechanism", The Euromicro Journal,
North-Holland, vol. 30, no. 1-5, (Aug.
1990), pp. 645-653.
[28] Refenes
A-P. N., "Parallelism in Knowledge Based Machines", The Knowledge
Engineering Review, Vol. 4 no.1, pp. 53-71 (1989).
[29] Refenes
A-P. N., & Alippi C., "Histological Image Understanding by Error
Backpropagation", Microprocessing and Microprogramming,
North-Holland, vol. 32, (1991) pp .437-446.
[30] Treleaven
P.C. and Refenes A-P. N., "Fifth Generation and VLSI Architectures",
North-Holland FGCS, vol. 1, no. 6, pp. 387-396, (Dec 1985).
[31] Eberbach
E., McCabe S. C., and Refenes A-P. N., "PARLE: a language for expressing
parallelism and integrating symbolic and numeric processing", The
Euromicro Journal, North-Holland, vol. 27, no. 1-5, (Sept. 1989), pp.
207-214.
[32] Refenes
A-P. N., Eberbach E., and Cotronis J., "Language Support for Concurrent
Symbolic and Numeric Systems", The Euromicro Journal, (June 1989).
[33] Balou A.,
and Refenes A-P. N, "Designing a parallel Object Oriented Compiler Target language", The Euromicro
Journal, North-Holland, vol. 30,
no. 1-5, (Aug. 1990), pp. 457-465.
[34] Refenes
A-P. N., and Balou A., "The Design and Implementation of VOOM: a parallel
Virtual Object-Oriented machine", Microprocessing and Microprogramming,
North-Holland, vol. 32, (1991) pp.
289-296.
[35] Refenes
A-P. N., Bentz Y., and Burgess A. N., "Neural Networks in Investment
Management", Journal of Finance and Communications, no. 8, April
(1994), 95-101.
[36] Refenes
A-P. N. “Neural
Model Identification, Variable Selection & Model Adequacy”, Comment on Serrano-Cinca
C., “Feedforward Neural Networks in the
Classification of Financial Information”,
European Journal of Finance, Vol. 3, No. 3, (1997), pp. 183-231.
ISSN 1351-847X.
[37] Refenes
A-P. N. and Burgess A. N. “Comment”, in Arthur W. B. et al, “Asset Pricing
Under heterogeneous Expectations”, Economic Notes by Banca dei Paschi di
Siena, Vol. 26, np. 2, pp. 331-336, (1997).
[38] Bentz Y.,
Refenes A-P. N. and Connor J. “Vos donnees ont-elles un sens? Les systemes
intelligents et adapatifs en gestion des actifs financiers”, la Revue du
Financier, No. 109 (1997).pp. 17-32, ISBN 0223-0143.
Conferences
I
(full paper submission, three referees,
rejection rates 70-75%)
[39] Burgess
A-P.N. & Refenes A. N. "A Principled Approach to Neural Network
Modelling of Financial Time Series", Proc. IEEE ICNN'95 Perth
Australia, (Nov. 1995), ISBN 0-7803-1182-5.
[40] Refenes
A-P. N., and Mitrelias C., "Network Pruning by Weight Variance",
Proc. NIPS'93 Denver Colorado, in Cowan J., Tesauro G., and Alspector
J., (ed), "Advances in Neural Information Processing", vol. 6, Morgan
Kaufmann, San Francisco (1994).
[41] Refenes
A-P. N., Zapranis A., and Azema-Barac M., "Stock Ranking: Neural
networks Versus Multiple linear
Regression", Proc. IEEE ICNN'93
San Francisco (March 28 - April s 1993).
[42] Refenes
A-P. N. "Optimizing Connectionist Datasets with ConSTrainer:", Proc.
2nd IEEE Symposium on Parallel & Distributed Processing, IEEE
Computer Society Press 2087, ISBN 0-8186-2087-TH0328-5/90/0000/0806, Dallas -
Texas, (Dec. 9-13 1990).
[43] Refenes
A-P. N., "CLS: An Adaptive Learning Procedure and Its Application to Time
Series Forecasting", Proc. IJCNN-91, Singapore, (Nov. 1991).
[44] Refenes
A-P. N., & Vithlani S. "Constructive Learning by Specialisation",
Proc. ICANN-1991, Elsevier Science Publishers, (Noth Holand), ed Kohonen
T.,(June 1991) pp. 923-929.
[45] Tuv E.,
& Refenes A-P. N., "Handling Malicous Vectors in Hetero-associative
Data Samples", Proc. IJCNN '93, Nagoya Japan (1993).
[46] Balou A.,
& Refenes A-P. N., "VOOM: a parallel Virtual Object-Oriented machine",
2nd IEEE Symposium on Parallel & Distributed Processing, Dallas -
Texas, IEEE Computer Society Press
2087, (Dec. 9 - 13 1990).
[47] Refenes
A-P. N., et al "PARLE: A Parallel Target Language for Integrating Symbolic
and Numeric Processing", Lecture Notes in Computer Science,
Springer-Verlag, Vol 365, pp. 181-198, Proc. PARLE-89, Eidhoven, The
Netherlands, June 12-16, 1989.
[48] Burgess A.
N., Bunn D. W. and Refenes A-P. N.
“Neural Networks with Error Feedback Terms for Financial Time Series
Modelling”, Proc. SNN’97, May 21, 1997, Amsterdam.
[49] Refenes
A-P. N. and Connor J. T., "Biasing Towards Integer Solutions ", in
Amari S. et al (eds), Proc.ICONIPS-96,
Springer, Singapore, (1996), pp. 681-689. ISBN 981-3083-03-4.
[50] Diamond
C., Shadbolt J, Azema-Barac M., and Refenes A-P. N. "Neural Networks for Tactical Asset Allocation in the Global Bonds
Markets", Proc. IEE Third International Conference on ANNS, (1993)
pp. 118-122, ISBN 0 85296 573 7.
[51] Refenes
A-P. N., et al "Currency
Exchange Rate Forecasting by Error Backpropagation", Proc. Int. Conf. on
System Sciences, HICCS-25, Kauai, HawaII, Jan. 7-10, (1992), ISBN
0-8186-2435-3.
[52] Refenes
A-P. N., et al "An Integrated
Neural Network System for Image Understanding", in "Machine vision
systems Integration in Industry", SPIE-90, vol 1386, Boston, (Nov.
1990).
[53] Eberbach
E., Cotronis J., and Refenes A-P. N., "Language Transformations for
concurrency control in symbolic and numeric systems", in Proc. Parallel
Computing 89, PARCOM-89, Leiden, The Netherlands, 29 Aug. - 1 Sept (1989),
in Evans D. J. et al (ed), ISBN:
0-44-88386X.
[54] Refenes
A-P. N., & Chan E. B., "Sound Recognition and Optimal Neural
Network Design", Proc. EUROMICRO-92,
Paris (Sept. 1992).
[55] Azema-Barac
M. M., & Refenes A-P. N., "Neural Network Implementations and Speed-up
on Massivelly Parallel Machines", Proc. EUROMICRO-92, Paris (Sept.
1992).
Book
chapters
[56] Skintzi
V., Skiadopoulos G., Refenes A-P. N. “The Effect of Misestimating Correlation
on Value-At-Risk”, in Refenes A-P. N. (ed), Quantitative Methods in Finance,
Typothito, Athens (2004), pp. 233-269, ISBN 960-402-173-7.
[57] Holt W.,
& Refenes A-P. N., “The D. W. for
Neural Models”, in Bol G. et al
(eds), “Risk Measurement, Econometrics, and Neural Networks”, Physica-Verlag,
Heidelberg, (1998), pp.57-69, ISBN 88/2202-5 43210.
[58] Refenes
A-P. N. Gonzalez F. and Burgess A. N., “A Principled Approach to Time Series
Analysis with Neural networks: An Application to Volatility Forecasting”,. In
Fiesler E., and Beale R. (eds) Handbook of Neural Computation, Oxford University
Press, (2002), pp F3.3-42.
[59] Bolland
P., Connor J. T. and Refenes A-P. N., “Application of Neural Network to
Forecast High Frequency Data”, in Dunis C. and Zhou B. (ed) “Non Linear
Modelling of High Frequency Financial Time Series”, Wiley & Sons, (1998),
pp. 225-246, ISBN 0-471-97464-1.
[60] Refenes
A-P. N., and Bolland P. "Modelling quarterly returns on the FTSE: A
comparative study with regression and neural networks", in Chen C. H. (ed)
" Fuzzy logic and Neural Network Handbook", Computer Engineering Series,
chapter 19, McGraw-Hill, N.Y. (1996), ISBN 0-07011189-8.
[61] Burgess A.
N., and Refenes A-P. N., "The Use of Error Feedback Terms in Neural
Network Modelling of Financial Time Series", in Dunis C. "Forecasting
Financial Markets", Wiley & Sons, Chichester (1996), pp. 261-275, ISBN
0-471-96653-3.
[62] Refenes
A-P. N. and Zapranis A. D., "Neural Networks in Tactical Asset Allocation:
a comparative study with Regression Models", in Arbib M. A. (ed),
"The Handbook of Brain Theory and Neural Networks", Bradford Books/The
MIT Press, 1995, ISBN 0-262-01148-4. Pp.491-495.
[63] Refenes
A-P. N., Zapranis A. D. Connor J. and Bunn D. W., "Neural Modelling in
Investment Management", in Treleaven P. C. and Goonatilake S. (ed),
"Intelligent Systems For Finance and Business", pp. 177-208, Wiley & Sons (1995), ISBN 0-471 94404 1.
[64] Refenes
A-P. N "Neural Networks in Investment Management: Testing Strategies and
Performance Metrics," in Zenios S., (ed), "Quantitative Methods,
Super computers and AI in Finance", pp. 287-308, Stanley Thornes (1995),
ISBN 0 7487 2336 6.
[65] Refenes
A-P. N. "Methods for Optimal Network Design", in Refenes A. N.
"Neural Networks in the Capital Markets", Wiley & Sons (1995).
[66] Refenes
A-P. N. "Data Modelling Considerations", in Refenes A. N.
"Neural Networks in the Capital Markets", Wiley & Sons (1995).
[67] Refenes
A-P. N., "Constructive Learning and its Application to Currency Exchange
Rate Prediction", in "Neural
Network Applications in Investment and Finance Services", Turban E., and
Trippi R. (eds), Chapter 27, Probus Publishing, Chicago, (1994).
[68] Refenes
A-P. N., Zapranis A. D., & Bentz Y., "Modeling Stock returns With
Neural Networks", in Lisboa P. G., and Taylor M, "Neural Networks -
II: techniques & Applications", Ellis Horwood (1994).
[69] Refenes
A-P. N., & Zaidi A., "Managing Exchange Rate Prediction Strategies
with Neural Networks", in Lisboa P. G., and Taylor M, "Techniques and
Applications of Neural Networks", Ellis Horwood (1993), ISBN 0 13 0 62183
8.
[70] Refenes
A-P. N., "ConSTrainer: A Generic Toolkit For Connectionist Dataset
Selection", in Dorffner G. (ed), "Konnektionismus in Artificial
Intelligence und Kognitionsforschung", Springer-Verlag, vol 252, (1990),
pp. 163-177.
[71] Refenes
A-P. N., "Parallel Abstract Machines: Towards a Unifying Intermediate Representation?"
in Warwick K. (ed) "Applied Artificial Intelligence", P.Peregrinus,
London, (1991) pp. 94-125. ISBN-0863412459.
[72] Treleaven
P.C., Refenes A-P.N., Lees K.J., McCabe S.C., "Computer Architectures for
Artificial Intelligence", in "Future Parallel Computers",
Trelevean P. C. and Vaneschi M., (eds), Springer-Verlag, Lecture Notes in
Computer Science, pp. 416-194, (Aug. 1987).
[73] Refenes
A-P. N. and Odjik E. A. M., "European Parallel Computing", in
"Parallel Computers; Object Oriented, Functional, Logic", Treleaven
P. C., (ed), Wiley & sons, London, (1989) pp.1-15.
[74] Refenes
A-P. N. and Bunn D. W., "Neural Networks and Investment Management",
in Skeete H., (ed), "The Handbook of World Stock and Commodity
Exchanges", IFR London (1995), ISBN 1 873446 65 9.
Conferences
II
(programme
committee review, rejection rates 55-60%)
[75]
Skintzi V. D. and Refenes A-P. N., "Volatility
spillovers and dynamic correlation in European Bond Karkets”, in Proc.,
International Bond & Debt Markets Integration Conference, Dublin,
31/05/2004.
[76]
Refenes A-P. N., and Holt W. T. "Non-Linear Models of
Financial Market Volatility”, in Proc., KES-2001, Knowledge Based Intelligent
Information Engineering Systems & Allied Technologies, Osaka, Japan (2001),
Baba N., Jain L. C. and Howlett R. J., (eds), IOS Press, ISBN 1-58603-192-9.
[77]
Refenes A-P. N., Zapranis A. D. and Utans J. "Model Identification, Variable
Selection and Model Adequacy ", Proc. NnCM' 96, Pasadena Nov.
19-21, (1996), also in Weigend A. et al
(ed) Neural Networks in Financial Engineering, World Scientific, Singapore,
(1997), pp. 243-262. ISBN 981-02-3123-7.
[78]
Zapranis A. D. Utans J. and Refenes A-P. N.,
"Specification Tests for Neural Networks: a case study in Tactical Asset
Allocation", Proc. NnCM' 96, Pasadena Nov. 19-21, (1996), also in
Weigend A. et al (ed) Neural Networks
in Financial Engineering, World Scientific, Singapore, (1997),pp 262-276. ISBN 981-02-3123-7.
[79]
Utans J. Holt W. and Refenes A-P. N. "Principle
Component Analysis for Modeling Multi-Currency Portfolios”, Proc. NnCM' 95,
London Oct. 11-13, (1995), also in Weigend A. et al (ed) Neural Networks in Financial Engineering, World
Scientific, Singapore, (1997), ISBN 981-02-3123-7.
[80]
Bentz Y., Refenes A-P. N., and De Laulanie J-F. "Neural
Network Meta-Models of Investment Strategy Performance", Proc. NnCM' 95,
London Oct. 11-13, (1995), also in Refenes A-P. N. et all (ed) Neural Networks
in Financial Engineering, World Scientific, Singapore, (1996), pp. 241-259,
ISBN 981-02-2480x.
[81]
Burgess A. N. and Refenes A-P. N., "Modelling
non-linear Co-integration in International Equity Index Futures", Proc. NnCM'
95, London Oct. 11-13, (1995), also in Refenes A-P. N. et all (ed) Neural
Networks in Financial Engineering, World Scientific, Singapore, (1996), pp.
50-64, ISBN 981-02-2480x.
[82]
Kollias C. and Refenes A-P. N. "Modelling the Effects
of Defense Spending Reductions using Neural Networks: Evidence from
Greece", Proc. ASSA'96, San Francisco (January 5-7, (1996).
[83]
Refenes A-P. N., "Measuring the Performance of Neural
Networks in Modern Portfolio Management", Proc. Unicom-95, in
Taylor J. G. (ed) Neural Networks, Alfred Waller, (1995), ISBN 1 872474276.
[84]
Refenes A-P. N., et al
"Financial Modelling Using Neural Networks", Proc. Unicom 92,
also in Liddell H. (ed) "Commercial Parallel Processing", Unicom,
(Feb. 1992).
[85]
Oliveira C. E. T., & Refenes A-P. N., "BROOM:
Designing a Parallel VLSI Architecture for Object-Oriented Systems", (IV
SBMICRO), Int. Conf. Microelectronics, Brazil, pp. 79-89, 12-14 July 1989.
[86]
Oliveira C. E. T., & Refenes A-P. N., "VLSI design
issues for a Basic Regular Object Oriented machine", IFIP workshop
on "Parallel Architectures on Silicon", Grenoble, (Nov. 1989).
[87]
Refenes A-P. N., and McKay S. C. "Parallel
Multi-Microcomputer Architecture to support the Integration of Symbolic and
Numeric Processing", Proc. Int. Conf. on Parallel Processing for Computer
Vision and Display, University of Leeds, (12-15 Jan. 1988).
[88]
Refenes A-P. N., et al,
"SPAN: Parallel Computer Systems for Integrated Symbolic and Numeric Processing",
in "ESPRIT-88: Putting the Technology to use", North-Holland,
pp.877-890, (Nov. 14-17, 1988).
others
(invited
journal/conference and/or refereed but non-ISBN publications)
[89]
Refenes A-P. N. and Bunn D. W., "Finance and Investment
Technology", Global Investment Management", Vol. 4:2, pp.
20-31, (1995).
[90]
Kollias C. and Refenes A-P. N. "Modelling the Effects
of Defence Spending Reductions on Investment Using Neural Networks in the Case
of Greece", Centre of Planning and Economic Research, Athens, TR-57/96,
(June 1996).
[91]
Refenes A-P. N., Bentz Y. Bunn W. D., Burgess A. N. and
Zapranis A. D, "Backpropagation with Discounted Least Squares and its
Application to Financial Time Series Modelling", Proc. NnCM' 94, Caltech
Pasadena, Nov. 17-18 (1994).
[92]
Refenes A-P. N., "Measuring the performance of Neural
Networks in modern Portfolio Management: Testing Strategies and Metrics",
Proc. NnCM' 94 Caltech Pasadena, Nov. 17-18 (1994)
[93]
Bentz Y., and Refenes A-P. N., "Backpropagation with
Weighted Signs and its Application to Financial Time Series", Proc. NnCM'
94 Caltech Pasadena, Nov. 17-18 (1994).
[94]
Zapranis A. D. and Refenes A-P. N., "Neural Networks in
Tactical Asset Allocation: Towards a Methodology for Hypothesis Testing and
Confidence Intervals:, Proc. NnCM' 94 Caltech Pasadena, Nov. 17-18 (1994).
[95]
Bolland P. J., and Refenes A-P. N., "Analysis o the
Relationship Between Volume, Open Interest and Futures Prices:, Proc. NnCM' 94
Caltech Pasadena, Nov. 17-18 (1994).
[96]
Refenes A-P. N., Bentz Y. Burgess A. N. and Zapranis A. D,
"Backpropagation with Differential Least Squares and its Application to
Financial Time Series Modelling", Proc. Snowbird 1994.
[97]
Refenes A-P. N. (ed), "Neural Networks in the Capital
Markets", Proc. NnCM-93 London Business School, Nov. 17-18 (1993).
[98]
Refenes A-P. N., Francis G., and Zapranis A., "Asset
Management within the APT Framework using Neural Networks", Proc. IFIP,
ORAIS'93: AI in Business and Finance, London (July 27 - 30 1993).
[99]
Refenes A-P. N. and Bilge U., "Sensitivity analysis for
Tactical Asset Allocation in the Global Bond Markets", in Refenes A. N,
(ed) "Neural Networks in the Capital Markets", Proc. NnCM'93 London
Business School, Nov. (1993).
[100]
Refenes A-P. N., Zapranis A. D., & Bentz Y.,
"Modeling Stock returns With Neural Networks", Proc. NnCM '93, London
Business School, Nov. 17-18 (1993).
[101]
Refenes A-P. N. and Bilge U., "Self-Organising Feature
maps in preprocessing datasets for decision support in Histopathology",
NSC/BME-90E, Proc. Int. North Sea Conf. in Biomedical Engineering, Antwerp, (19-22 Nov. 1990).
[102]
Refenes A-P. N., Kollias, C., and Zapranis A., "Arms
Race Modeling using Neural Networks: a case study", Proc. Yapay Zeka
Sempozyum, Istanbul, (June 1993).
[103]
Uchida S., and Refenes A-P. N. (eds) "Benchmarking for
Parallel Systems", Proc. 2nd Joint Workshop ICOT/DTI-SERC, Tokyo, (Oct.
15-17, 1990).
[104]
Refenes A-P. N., "Neurocomputing: Key Requirements on
Basic Theory & Enabling Technology", Proc NIPT' 91, Tokyo, (March
13-14, 1991).
[105]
Refenes A-P. N., "NeuroComputing: Themes and
variations", Proc. Int. Workshop, Samarkand, Uzbekistan, USSR, (Sept.
10-17, 1990).
[106]
Refenes A-P. N., "Parallel Computing in Europe and the
UK", Proc. 2nd Joint Workshop ICOT/DTI-SERC, Tokyo, (Oct. 15-17, 1990).
[107]
Refenes A-P. N., "Parallel Knowledge Processing
Computers", Proc. BCS Symp. on "Parallel Architectures for Artificial
Intelligence", Birkbeck College London, (Feb-23 1988), pp. 1-20.
[108]
Refenes A-P. N., "Neural Computing Technology Transfer
Programme", Workplan & Strategy Report, Department of Trade and
Industry, Information Technology Division (Aug. 1991).
[109]
Refenes A-P. N., "Neural Network Design Strategies for
Histological Image Understanding", Proc. Applications of Neural Computing
in Medicine, Institute of Physical Sciences in Medicine, Royal Marsden
Hospital, (April 1992), London.
under
review/pending
[110]
Zapranis A. D., Utans J. and Refenes A-P. N. “Sampling
Variability Estimation Schemes For Neural Models”, Neural Networks
(submitted).
[111]
Holt W. and Refenes A-P. N. “Two residual diagnostic test
statistics for neural regression models”, ISAS ’98, 4th Int. Conf.
On Information Systems Analysis and Synthesis, Orlando-Florida July 12-16 1998
(submitted March 1998, Accepted June 1998).
[112]
Holt W. and Refenes A-P. N. “A principled approach to neural
regression analysis: a case study on the airline data”, Applied Statistics,
Journal of the Royal Statistical Society (submitted, under review).