Skip to main content

Applications of Computational Intelligence in Data-Driven Trading

Hardcover

Pre-order

£47.50

*VAT

Applications of Computational Intelligence in Data-Driven Trading

Cris Doloc

ISBN: 978-1-119-55050-1 October 2019 256 Pages

Description

An introduction to applications of computational intelligence in finance

Applications of Computational Intelligence in Data-Driven Trading features modern educational content that is at the confluence between data-driven decision-making and computational intelligence.

The book caters to trading and investment professionals interested in the new paradigm of data-driven decision-making, as well as to graduate students who desire to get more familiar with the emerging field of computational intelligence. 

Doloc introduces the reader to the new paradigm of Data-Intensive Computing and its applications in the world of trading and investing. The goal is to promote the use of computational intelligencetechniques, as the vehicle to augment human performance thorough automation and emulate human intelligence via innovation and discovery. Several case studies from the field of data-driven trading and investing are presented.

The author’s two decades of experience as a computational scientist and quantitative practitioner in the financial trading industry endow him with a unique perspective that he conveys to the reader. The financial trading industry is fertile ground for the adoption of advanced technologies, and Doloc walks the reader through two key areas: automation and innovation.

About the Author

Acknowledgments

About the Website

Introduction

Motivation

Target audience

Book structure

Acknowledgements

1 The evolution of trading paradigms

1.1 Infrastructure-related paradigms in Trading

1.1.1 The open outcry trading

1.1.2 Advances in communication technology

1.1.3 The Digital revolution in the financial markets

1.1.4 The High Frequency Trading paradigm

1.1.5 Blockchain and the decentralization of markets

1.2 Decision-Making paradigms in trading

1.2.1 Discretionary trading

1.2.2 Algorithmic trading

1.3 The new paradigm of Data-Driven Trading

References

2 The role of Data in Trading and Investing

2.1 The Data-driven decision-making paradigm

2.2 The Data economy is fulling the future

2.3 Defining Data and its utility

2.4 The journey from Data to Intelligence

2.5 The utility of Data in Trading and Investing

2.5.1 The use of Big Data analytics to feed financial models

2.5.2 The use of real-time analytics

2.5.3 The use of Machine Learning

2.5.4 Automated Risk Management

2.5.5 Data management

2.5.6 Consumer analytic

2.5.7 Fraud detection

2.6 The Alternative data and its use in Trading and Investing

References

3 Artificial Intelligence – between myth and reality

3.1 Introduction

3.2 A brief history of AI

3.2.1 Early history

3.2.2 The modern AI era

3.2.3 Important milestones in the development of AI

3.2.4 Projections for the immediate future

3.2.5 Meta-Learning – an exciting new development

3.3 The meaning of the term “AI” – a critical view

3.4 On the applicability of “AI” to Finance

3.4.1 Data stationarity

3.4.2 Data quality

3.4.3 Data dimensionality

3.5 Perspectives and future directions

References

4 Computational Intelligence: A principled approach for the era of Data exploration

4.1 Introduction to Computational Intelligence

4.1.1 Defining Intelligence

4.1.2 What is Computational Intelligence?

4.1.3 Mapping the field of study

4.1.4 Problems vs. tools

4.1.5 Current challenges

4.1.6 The future of Computational Intelligence

4.1.7 Examples in Finance

4.2 The PAC theory

4.2.1 The Probably Approximately Correct framework

4.2.2 Why AI is a very lofty goal to achieve?

4.2.3 Examples of Ecorithms in Finance

4.3 Technology drivers behind the ML surge

4.3.1 Data

4.3.2 Algorithms

4.3.3 Hardware Accelerators

References

5 How to apply the principles of CI in Quantitative finance

5.1 The viability of Computational Intelligence

5.2 On the applicability of CI to Quantitative finance

5.3 A brief introduction to Reinforcement Learning

5.3.1 Defining the Agent

5.3.2 Model-based Markov Decision Process

5.3.3 Model-free Reinforcement Learning

5.4 Conclusions

References

6 Case Study 1: Optimizing trade execution

6.1. Introduction to the problem

6.1.1 On Limit Orders and market Microstructure

6.1.2 Formulation of base-line strategies

6.1.3 A RL formulation of the Optimized Execution problem

6.2 Current State-of-the-Art in Optimized Trade execution

6.3 Implementation methodology

6.3.1 Simulating the interaction with the market Microstructure

6.3.2 Using Dynamic Programming to optimize trade execution

6.3.3 Using Reinforcement Learning to optimize trade execution

6.4 Empirical results

6.4.1 Application to Equities

6.4.2 Using “private” variables only

6.4.3 Using both “private” and “market” variables

6.4.4 Application to Futures

6.4.5 Another example

6.5 Conclusions and future directions

References

7 Case Study 2: The Dynamics of the Limit Order Book

7.1. Introduction to the problem

7.1.1 The new Era of Prediction

7.1.2 New Challenges

7.1.3 High Frequency data

7.2 Current SOTA in the prediction of directional price movement in the LOB

7.3 Using SVM and RF classifiers for directional price forecast

7.4 Studying the dynamics of the LOB with RL

7.5 Studying the dynamics of the LOB with DNN

7.6 Studying the dynamics of the LOB with LSTM

7.7 Studying the dynamics of the LOB with CNN

References

8 Case Study 3: Applying ML to Portfolio Management

8.1 Introduction to the problem

8.2 Current State-of-the-Art in Portfolio modelling

8.2.1 The classic approach

8.2.2 The ML approach

8.3 A “Deep Portfolio” approach to portfolio optimization

8.3.1 Auto-encoders

8.3.2 Methodology – the 4-step algorithm

8.3.3 Results

8.4 A Q-learning approach to the problem of portfolio optimization

8.4.1 Problem statement

8.4.2 Methodology

8.4.3 The Deep Q-learning algorithm

8.4.4 Results

8.5 A Deep RL approach to portfolio management

8.5.1 Methodology

8.5.2 Data

8.5.3 The RL setting: agent, environment and policy

8.5.4 The CNN implementation

8.5.5 The RNN and LSTM implementations

8.5.6 Results

References

9 Case Study 4: Applying ML to Market Making

9.1 Introduction to the problem

9.2 Current State-of-the-Art in Market Making

9.3 Applications of Temporal-Difference RL in Market Making

9.3.1 Methodology

9.3.2 The Simulator

9.3.3 Market Making Agent specification

9.3.4 Empirical Results

9.4 Market Making in HFT using RL

9.4.1 Methodology

9.4.2 Experimental setting

9.4.3 Results and Conclusions

9.5 Other research studies

References

10 Case Study 5: Applications of ML to Derivatives valuation

10.1 Introduction to the problem

10.2 Current State-of-the-Art in Derivatives valuation by applying ML

10.2.1 The beginnings

10.2.2 The last decade

10.3 Using Deep Learning for valuation of Derivatives

10.3.1 Implementation Methodology

10.3.2 Empirical Results

10.3.3 Conclusions and future directions

10.3.3 Other research studies

10.4 Using Reinforcement Learning for valuation of Derivatives

References

11 Case Study 6: Using ML for Risk Management and Compliance

11.1 Introduction to the problem

11.2 Current State-of-the-Art for applications of ML to Risk Management and Compliance

11.2.1 Credit risk

11.2.2 Market risk

11.2.3 Operational risk

11.2.4 Regulatory Compliance risk

11.3 ML in Credit Risk modelling

11.3.1 Data

11.3.2 Models

11.3.3 Results

11.4 Deep Learning for Credit scoring

11.3.1 Credit risk

11.4.2 Deep Belief networks and RBM

11.4.3 Empirical Results

11.5 Using ML in Operational Risk and Market surveillance

11.5.1 Intro

11.5.2 A ML approach to market surveillance

11.5.3 Conclusions

References

12 Conclusions and future directions

12.1 Concluding remarks

12.2 The Paradigm shift

12.3 De-noising the AI hype

12.4 The birth of a new Engineering discipline

12.5 Future directions

References