Motivation:There is an explosive growth of Algorithmic Trading (i.e., Algo Trading, Program Trading, or Automated Trading) in research and practice over the past few years. As the markets are evolving fast, numerous problems have arisen in the field of Algorithmic Trading due to many reasons. Just a few of them are mentioned below.
First of all, as high volume and high variety heterogeneous data at different frequency, e.g., high and even ultra-high frequency, are exploding in the market, the modeling tasks to explore the big information across from structure to non-structure data have become increasingly complex. Second, with different policy manipulation of market, the market regimes have changed fast in volatility. The demand to achieve robust strategies has even been strong. Finally, as shown in the work by 2013 Nobel laureates, Fama and Shiller, the market is a changing mixture of efficiency and “irrational exuberance”. It has been interesting and extremely challenging for an investment organization to balance the investment horizons for its portfolio and select different trading strategies to explore and exploit inefficiency and irrationality in the market and then make profits and control risks.
All these kinds of problems attract professionals and researchers to intensively study novel methodology with advanced tools. In many cases, conventional mathematical approaches do not well support the automated trading. Computational/artificial intelligence has the power in adaptively learning the models from data, inferring the market states upon the new information and naturally accounting for the uncertainty. We believe that advanced computational intelligence approaches will mitigate and even someday solve the existing and emerging problems in our trading practice by helping us build up intelligent trading agents.
Goal:This special session is an approved plan by IEEE CIS Computational Finance and Economics Technical Committee (CFETC) .
It aims to bring practical pioneers and academic researcher together, provide an idea exchanging environment, and explore potential collaboration between industry and academia. In this session, we will explore new theories and solve real trading problems. We also hope to re-exam the Algorithmic Trading state-of-the-art and paradigms under recent data complexity development and policy risks.
Topics:Topics of interests include, but not limited to, trading models and strategies, risk management, pricing for algorithmic trading, and strategy validation for different underlines in different markets, as follows:
- Connectionist approaches, e.g., neural networks, for learning and approximating price-related functions, market prediction and other novel applications.
- Deep and shallow learning of market structure.
- Bayesian approaches for modeling market factors.
- Non-parametric statistical approaches for trading activities.
- Latent and hidden structure methods for market regime and state identification.
- Transfer learning for information borrowing from different markets and assets.
- Adaptive learning/control for achieving robust strategies.
- Reinforcement learning paradigm for handling and decision-making under high uncertainty.
- Agent-based models and their applications in artificial market and other directions.
- Behavior-based approaches for understanding market inefficiency and irrationality.
- Trading models/agents and strategies evolution.
- Fuzzy system combination with other approaches for inference and decision-making.
- Information theoretic methods and other approaches for portfolio optimization and for optimally allocating capital between trading strategies.
- Utilization of non-structure data and big data in trading practice.
- Validation of algorithmic trading models and strategies.
- Better understanding and control of risk during trading with advanced intelligent modeling.
Information for AuthorsThis section is part of IEEE International Joint Conference on Neural Network 2014 (IEEE IJCNN 2014) at The IEEE World Congress on Computational Intelligence 2014 (IEEE WCCI 2014).
1) Information on the format and templates for papers can be found here:
2) Papers should be submitted via the IJCNN 2014 paper submission site:
3) Select the Special Session name in the Main Research topic dropdown list
4) Fill out the input fields, upload the PDF file of your paper and finalize your submission by the deadline of December 20, 2013
Important datesPaper submission: 20 December, 2013
Decision: 15 March, 2014
Final paper submission: 15 April, 2014
Conference dates: 6-11 July, 2014
Organizers:Dr. Chenghui Cai
Global Market Department, Opera Solutions LLC, New York/New Jersey.
Dr. Ming Li
Taikang Asset Management Company, Beijing and Taikang Financial Engineering Department.
Dr. Meng Ji
Ernst & Young, New York City.
Prof. Akira Namatame
Department of Computer Science, the Japan Defense Academy (NDA).
Prof. Philip Yu
Department of Statistics and Actuarial Science of The University of Hong Kong.
Prof. Kiyoshi Izumi
School of Engineering, The University of Tokyo
Mr. Robert Golan
DBmind Technologies, USA.