ICONIP'96 -- Tutorial Information

TUTORIAL ON FINANCIAL ENGINEERING

Three tutorials on financial application of Neural Networks will be presented on Tuesday, September 24, 1996, which is one of the events of the 1996 International Conference on Neural Information Processing (ICONIP'96) to be held from 24-27 September, 1996. Each tutorial lasts for 3 hours, including a 15 minutes tea break and will be held in the Hong Kong Convention and Exhibition Centre.

Schedule:

8:45am -- 11:45am	Tutorial 1: "Time Series Models: Classical and Nonlinear Approaches", by Prof. John E Moody, Oregon Graduate Institute.
11:45am -- 12:45pm	Lunch Break
12:45pm -- 3:45pm	Tutorial 2: "Neural Networks in Financial Engineering", by Prof. Paul Refenes, London Business School.
3:45pm -- 4:00pm	Tea Break
4:00pm -- 7:00pm	Tutorial 3: "Option Pricing in Modern Finance Theory and the Relevance of Artificial Neural Network", by Prof. Halbert White, Univ. of Calif., San Deigo.

Registration Information
Further Information
Registration form
Download a PostScript Version of the form

Tutorial 1: Time Series Models: Classical and Nonlinear Approaches.
Professor John E. Moody
Computer Science Department, Oregon Graduate Institute, U.S.A.

This tutorial will present an overview of both classical and nonlinear approaches to analyzing and predicting time series. The presentation will begin with a review of linear time series models (eg. AR, MA, ARIMAX), followed by simple nonlinear models (eg. bilinear, regime switching, ARCH). Relevant results of chaos theory, including the limits of predictability, strange attractors, and "embeddology" will be summarized. Next, general nonlinear models, including both feed forward and recurrent neural networks will be discussed. Finally, I will present several time series modeling case studies from economics, physics, and medicine will be presented.

Tutorial 2: Neural Networks in Financial Engineering
Professor A-P. N. Refenes
Neuro Forecasting Unit, London Business School, U.K.

The potential for computational intelligence applications in financial engineering has yet to fully take off. For neural networks to become common place in the financial engineering community, it requires the synergetic combination of expertise in financial economics and the mathematical/natural sciences. The objectives of this short course are as follows:

to create an awareness in the neural network and engineering communities of the basic needs and approaches of financial engineering.
to formulate neural learning in terms of econometric and statistical model which will expose the fundamental advantages and computational properties of the technology
to create an awareness of the breadth of applicability of neural techniques in diverse areas of financial engineering.
to expose some of the basic weaknesses of neural modelling techniques when applied to finance problems in which the stochastic component in the data generating processes is by far the most prevalent, in comparison to any deterministic component.
to identify the main areas of the research agenda if neural networks are to become commonplace in finance and econometrics alongside their statistical counterparts.

The main elements in the programme reflect these objectives. This is an introductory tutorial, a collection of case studies, and a session on the main research topics that need to be dealt with to satisfy the last objective (amongst others).

Tutorial 3: Option Pricing in Modern Finance Theory and
the Relevance of Artificial Neural Networks
Professor Halbert White
Department of Economics, University of California, San Diego, U.S.A.

This tutorial will review the theory of option pricing in the modern mathematical finance literature, including the popular Black-Scholes model and some of its extensions. With this as foundation, we will investigate the role that artificial neural networks can play in pricing options and in fashioning appropriate option trading strategies. The tutorial will assume a basic knowledge of probability theory and analysis.

Registration Fee:

	One Tutorial	All 3 tutorials
Member*	HK$ 1,000	HK$ 1,500
Non-member	HK$ 1,200>	HK$ 1,800
Students**	HK$ 1,000>	HK$ 1,500

* You are eligible to the member rate if you are a member of IEEE, ACM, HKCS, CNNC, ENNS, INNS or JNNS. Photocopies of your membership card must be provided with the application form.
** Photocopies of student cards must be provided together with the application form

Registration fee include tutorial, tutorial notes and coffee breaks. Lunch is NOT inlcuded.

Completed form should be sent to:
ICONIP'96 Secretariat,
Department of Computer Science & Engineering,
The Chinese University of Hong Kong,
Shatin, N.T. Hong Kong.

All Cheques should be made payable to ICONIP'96

For further information, please contact:
Dr K P Chan,
Department of Computer Science,
The University of Hong Kong.
e-mail: kpchan@cs.hku.hk
Fax: (+852) 2559 8447

SPONSORS OF ICONIP'96