Table of Contents
List of figures xiii
List of tables xvii
Notes on contributors xix
1 Machine learning in finance and accounting Mohammad Zoynul Abedin M. Kabir Hassan Petr Hajek Mohammed Mohi Uddin 1
1.1 Introduction 1
1.2 Motivation 2
1.3 Brief overview of chapters 3
References 4
2 Decision trees and random forests Roberto Casarin Alessandro Facchinetti Domenico Sorice Stefano Tonellato 7
2.1 Introduction 7
2.2 Classification trees 8
2.2.1 Impurity and binary splitting 9
2.2.1.1 Specification of the impurity function 10
2.2.1.2 Labeling the leaves 11
2.2.1.3 Tree size and stopping rules 12
2.2.2 Performance estimation 12
2.2.2.1 Resubstitution estimate 13
2.2.2.2 Test-sample estimate 13
2.3 Regression trees 14
2.3.1 Regression 14
2.3.2 Performance assessment and optimal size of the tree 15
2.3.2.1 Resubstitution estimate of MSE(T) 15
2.3.2.2 Test-sample estimate of MSE(T) 15
2.4 Issues common to classification and regression trees 16
2.4.1 Surrogate splits 16
2.4.1.1 Handling of missing values 17
2.4.1.2 Ranking of input variables 18
2.4.1.3 Input combination 18
2.4.2 Advantages and disadvantages of decision trees 18
2.5 Random forests 19
2.5.1 Prediction error bias-variance decomposition 19
2.5.2 Bias-variance decomposition for randomized trees ensembles 21
2.5.3 From trees ensembles to random forests 22
2.5.4 Partial dependence function 23
2.6 Forecasting bond returns using macroeconomic variables 24
2.7 Default prediction based on accountancy data 28
2.8 Appendix: R source codes for the applications in this chapter 30
2.8.1 Application to US BofA index 31
2.8.2 SME default risk application 34
References 35
3 Improving longevity risk management through machine learning Susanna Levantesi Andrea Nigri Gabriella Piscopo 37
3.1 Introduction 37
3.2 The mortality models 39
3.3 Modeling mortality with machine learning 41
3.4 Numerical application 43
3.4.1 Mortality models by comparison: an empirical analysis 43
3.4.2 Longevity management for life insurance: sample cases 46
3.5 Conclusions 48
3.6 Appendix 49
Note 55
References 55
4 Kernel switching ridge regression in business intelligence systems MD. Ashad Alam Osamu Komori MD. Ferdush Rahman 57
4.1 Introduction 57
4.2 Method 59
4.2.1 Switching regression 59
4.2.2 Switching ridge regression 60
4.2.3 Dual form of the ridge regression 60
4.2.4 Basic notion of kernel methods 61
4.2.5 Alternative derivation to use ridge regression in the feature space 61
4.2.6 Kernel ridge regression 62
4.2.7 Kernel ridge regression: duality 63
4.2.8 Kernel switching ridge regression 65
4.3 Experimental results 66
4.3.1 Simulation 66
4.3.2 Application in business intelligence 67
4.4 Discussion 70
4.5 Conclusion and future research 70
4.6 Appendix: Kernel switching ridge regression: an R code 71
References 72
5 Predicting stock return volatility using sentiment analysis of corporate annual reports Petr Hajek Renata Myskova Vladimir Olej 75
5.1 Introduction 75
5.2 Related literature 76
5.3 Research methodology 78
5.3.1 Financial data and indicators 79
5.3.2 Textual data and linguistic indicators 80
5.3.3 Machine learning methods 81
5.4 Experimental results 86
5.5 Conclusions 93
Acknowledgments 93
References 93
6 Random projection methods in economics and finance Roberto Casarin Veronica Veggente 97
6.1 Introduction 97
6.2 Dimensionality reduction 100
6.2.1 Principal component analysis (PCA) 101
6.2.2 Factor analysis 102
6.2.3 Projection pursuit 103
6.3 Random projection 103
6.3.1 Johnson-Lindenstrauss lemma 104
6.3.2 Projection matrices' specification 105
6.4 Applications of random projection 106
6.4.1 A compressed linear regression model 106
6.4.2 Tracking the S&P500 index 108
6.4.3 Forecasting S&P500 returns 111
6.4.4 Forecasting energy trading volumes 114
6.5 Appendix: Matlab code 118
Notes 120
References 120
7 The future of cloud computing in financial services: a machine learning and artificial intelligence perspective Richard L. Harmon Andrew Psaltis 123
7.1 Introduction 123
7.2 The role of machine learning and artificial intelligence in financial services 124
7.3 The enterprise data cloud 126
7.4 Data contextuality: machine learning-based entity analytics across the enterprise 127
7.5 Identifying Central Counterparty (CCP) risk using ABM simulations 131
7.6 Systemic risk and cloud concentration risk exposures 134
7.7 How should regulators address these challenges? 137
Notes 137
References 138
8 Prospects and challenges of using artificial intelligence in the audit process Emon Kalyan Chowdhury 139
8.1 Introduction 139
8.1.1 Background and relevant aspect of auditing 140
8.2 Literature review 141
8.3 Artificial intelligence in auditing 142
8.3.1 Artificial intelligence 142
8.3.2 Use of expert systems in auditing 143
8.3.3 Use of neural network in auditing 143
8.4 Framework for including AI in auditing 143
8.4.1 Components 144
8.4.1.1 AI strategy 144
8.4.1.2 Governance 144
8.4.1.3 Human factor 144
8.4.2 Elements 145
8.4.2.1 Cyber resilience 145
8.4.2.2 AI competencies 145
8.4.2.3 Data quality 145
8.4.2.4 Data architecture and infrastructure 145
8.4.2.5 Measuring performance 145
8.4.2.6 Ethics 145
8.4.2.7 Black box 146
8.5 Transformation of the audit process 146
8.5.1 Impact of digitalization on audit quality 147
8.5.2 Impact of digitalization on audit firms 147
8.5.3 Steps to transform manual audit operations to AI-based 148
8.6 Applications of artificial intelligence in auditing - few examples 149
8.6.1 KPMG 149
8.6.2 Deloitte 149
8.6.3 PwC 149
8.6.4 Ernst and Young (EY) 150
8.6.5 K. Coe Isom 150
8.6.6 Doeren Mayhew 150
8.6.7 CohnReznick 150
8.6.8 The Association of Certified Fraud Examiners (ACFE) 150
8.7 Prospects of an AI-based audit process in Bangladesh 150
8.7.1 General aspects 151
8.7.2 Audit firm specific aspects 151
8.7.3 Business organization aspects 152
8.8 Conclusion 152
Bibliography 153
9 Web usage analysis: pillar 3 information assessment in turbulent times Anna Pilkova Michal Munk Petra Blazekova Lubomir Benko 157
9.1 Introduction 157
9.2 Related work 158
9.3 Research, methodology 161
9.4 Results 164
9.5 Discussion and conclusion 172
Acknowledgements 175
Disclosure statement 175
References 175
10 Machine learning in the fields of accounting, economics and finance: the emergence of new strategies Maha Radwan Salma Drissi Silvana Secinaro 181
10.1 Introduction 181
10.2 General overview on machine learning 182
10.3 Data analysis process and main algorithms used 183
10.3.1 Supervised models 184
10.3.2 Unsupervised models 186
10.3.3 Semi-supervised models 187
10.3.4 Reinforcement learning models 188
10.4 Machine learning uses: cases in the fields of economics, finance and accounting 189
10.4.1 Algorithmic trading 189
10.4.2 Insurance pricing 190
10.4.3 Credit risk assessment 191
10.4.4 Financial fraud detection 192
10.5 Conclusions 194
References 194
11 Handling class imbalance data in business domain MD. Shajalal Mohammad Zoynul Abedin Mohammed Mohi Uddin 199
11.1 Introduction 199
11.2 Data imbalance problem 200
11.3 Balancing techniques 201
11.3.1 Random sampling-based method 201
11.3.2 SMOTE oversampling 201
11.3.3 Borderline-SMOTE 202
11.3.4 Class weight boosting 203
11.4 Evaluation metrics 203
11.5 Case study: credit card fraud detection 206
11.6 Conclusion 208
References 208
12 Artificial intelligence (AI) in recruiting talents: recruiters' intention and actual use of AI MD. Aftab Uddin Mohammad Sarwar Alam MD. Kaosar Hossain Tarikul Islam MD. Shah Azizul Hoque 211
12.1 Introduction 211
12.2 Theory and hypothesis development 213
12.2.1 Technology anxiety and intentions to use 214
12.2.2 Performance expectancy and intentions to use 214
12.2.3 Effort expectancy and intentions to use 214
12.2.4 Social influence and intention to use 215
12.2.5 Resistance to change and intentions to use 215
12.2.6 Facilitating conditions and intentions to use 215
12.2.7 Behavioral intention to use and actual use 216
12.2.8 Moderating effects of age status 216
12.3 Research design 218
12.3.1 Survey design 218
12.3.2 Data collection procedure and participants' information 218
12.3.3 Measurement tools 218
12.3.4 Results and hypotheses testing 219
12.3.4.1 Analytical technique 219
12.3.4.2 Measurement model evaluation 219
12.3.4.3 Structural model evaluation 221
12.3.4.4 Testing of direct effects 222
12.3.4.5 Testing of moderating effects 222
12.4 Discussion and conclusion 223
12.4.1 Limitation of study and future research directions 225
References 226
Index 233
