Financial Data Analytics: Theory and Application (Contributions to Finance and Accounting)

✍ Scribed by Sinem Derindere Köseoğlu (editor)

• Publisher: Springer
• Year: 2022
• Tongue: English
• Leaves: 393
• Category: Library

✦ Synopsis

​This book presents both theory of financial data analytics, as well as comprehensive insights into the application of financial data analytics techniques in real financial world situations. It offers solutions on how to logically analyze the enormous amount of structured and unstructured data generated every moment in the finance sector. This data can be used by companies, organizations, and investors to create strategies, as the finance sector rapidly moves towards data-driven optimization.
This book provides an efficient resource, addressing all applications of data analytics in the finance sector. International experts from around the globe cover the most important subjects in finance, including data processing, knowledge management, machine learning models, data modeling, visualization, optimization for financial problems, financial econometrics, financial time series analysis, project management, and decision making. The authors provide empirical evidence as examples of specific topics. By combining both applications and theory, the book offers a holistic approach.
Therefore, it is a must-read for researchers and scholars of financial economics and finance, as well as practitioners interested in a better understanding of financial data analytics.

✦ Table of Contents

Preface
About the Book
Financial Data Analytics: Theory and Application
Contents
Editor and Contributors
Part I: Introduction and Analytics Models
Retraining and Reskilling Financial Participators in the Digital Age
1 Introduction
2 Background
2.1 The Development of FinTech (Financial Technology) and Data Analytics
2.2 Reskilling and Retraining for Financial Professionals
3 What Knowledge and Skills Do Financial Professionals Are Needed in the Digital Age?
3.1 Emerging Hard Skills for Financial Professionals
3.2 Soft Skills for Financial Professionals
3.3 Personality Type and Career Choices in Financial Industry
4 Learning with Technology in Digital Age
4.1 Multimedia
4.2 Collaborative Technologies
4.3 Artificial Intelligence
5 The Challenge of Learning in the Big Data World
5.1 Challenges of Learning with Technologies
5.2 Learner-Related Challenges
5.3 Technological-Related Challenges
5.4 The Age of Big Data
6 How Do People Learn?
6.1 Related Theories of Learning
7 Solutions and Recommendations
8 Conclusion
References
Basics of Financial Data Analytics
1 Basic Terminology of Data Science
1.1 Classifications of Data
1.2 Data Analytics Types and Data Modelling
2 Data Science Process and Descriptive Analytics
2.1 Financial Time Series Characteristics and R
2.1.1 Volatility and Extreme Values of Asset Returns
2.1.2 Distribution Characteristics of Asset Returns
2.1.3 Stationarity Characteristics of Financial Time Series
2.1.4 Dependence (Autocorrelation/Serial Correlation/Serial Dependence/Mean Reversion)
2.2 Smoothing
3 Summary
References
Predictive Analytics Techniques: Theory and Applications in Finance
1 Introduction
2 Background
3 Main Focus of the Chapter
3.1 On Predictive Analytics
3.1.1 Programming in R
4 Solutions and Recommendations
4.1 Predictive Model 1: Logistic Regression
4.1.1 Foundation
4.1.2 Advanced Organizer
4.1.3 Assumptions
4.1.4 Objective
4.1.5 Data Source
4.1.6 Data Exploration
4.1.7 Predictive Tasks
4.1.8 Data Pre-processing
4.1.9 Build the Predictive Model
4.1.10 Model Performance Evaluation
4.1.11 Model Validation
4.1.12 Make Predictions
4.1.13 Validate Prediction
4.1.14 Results Interpretation
4.1.15 Conclusion
4.2 Predictive Model 2: Time Series Analysis
4.2.1 Foundation
4.2.2 Advance Organizer
4.2.3 Assumptions
4.2.4 Objective
4.2.5 Data Source
4.2.6 Predictive Tasks
4.2.7 Data Exploration and Pre-processing
4.2.8 Build the Predictive Model
4.2.9 Results Interpretation (of the Holt-Winters Output)
4.2.10 Conclusion
4.3 Predictive Model 3: Decision Tree
4.3.1 Foundation
4.3.2 Advance Organizer
4.3.3 Objective
4.3.4 Data Source
4.3.5 Data Exploration
4.3.6 Predictive Tasks
4.3.7 Data Pre-processing
4.3.8 Build the Predictive Model
4.3.9 Model Visualization
4.3.10 Making Predictions
4.3.11 Results Interpretation
4.3.12 Conclusion
4.4 Predictive Model 4: Multiple Linear Regression
4.4.1 Theoretical Foundation
4.4.2 Advance Organizer
4.4.3 Objective
4.4.4 Data Source
4.4.5 Data Exploration and Cleaning
4.4.6 Predictive Tasks
4.4.7 Data Pre-processing
4.4.8 Build the Predictive Model
4.4.9 Initial Verification of the Model
4.4.10 Further Model Validation
4.4.11 Results Interpretation
4.4.12 Conclusion
4.5 Predictive Model 5: RFM Segmentation with k-means
4.5.1 Foundation
4.5.2 Advance Organizer
4.5.3 Objective
4.5.4 Data Source
4.5.5 Data Exploration and Cleaning
4.5.6 Predictive Tasks
4.5.7 Data Pre-processing
4.5.8 Build RFM Model
4.5.9 Build the k-Means Model
4.5.10 Results Visualization
4.5.11 Results Interpretation
4.5.12 Conclusion
5 Future Research Directions
6 Conclusion
Key Terms and Definitions
References
Prescriptive Analytics Techniques: Theory and Applications in Finance
1 Introduction
2 Background
3 Main Focus of the Chapter
3.1 On Prescriptive Analytics
3.1.1 Programming in R
4 Solutions and Recommendations
4.1 Prescriptive Model 1: Sentiment Analysis
4.1.1 Foundation
4.1.2 Advance Organizer
4.1.3 Objective
4.1.4 Data Source
4.1.5 Data Exploration and Cleaning
4.1.6 Prescriptive Tasks
4.1.7 Build the Prescriptive Model
4.1.8 Results Visualization
4.1.9 Results Interpretation
4.1.10 Conclusion
4.2 Prescriptive Model 2: Association Rules
4.2.1 Foundation
4.2.2 Advance Organizer
4.2.3 Objective
4.2.4 Data Source
4.2.5 Data Exploration and Cleaning
4.2.6 Prescriptive Tasks
4.2.7 Build the Prescriptive Model
4.2.8 Results Interpretation
4.2.9 Conclusion
4.3 Prescriptive Model 3: Network Analysis
4.3.1 Foundation
4.3.2 Advance Organizer
4.3.3 Objective
4.3.4 Data Source
4.3.5 Data Exploration
4.3.6 Prescriptive Task
4.3.7 Build the Prescriptive Model and Interpretation
4.3.8 Conclusion
4.4 Prescriptive Model 4: Recommender Systems
4.4.1 Theoretical Foundation
4.4.2 Advance Organizer
4.4.3 Objective
4.4.4 Data Source
4.4.5 Data Exploration and Cleaning
4.4.6 Prescriptive Tasks
4.4.7 Data Pre-processing
4.4.8 Build the Prescriptive Model
4.4.9 Prescriptive Actionable Information
4.4.10 Conclusion
4.5 Prescriptive Model 5: Principal Components Analysis
4.5.1 Theoretical Foundation
4.5.2 Advance Organizer
4.5.3 Objective of the Model
4.5.4 Data Source
4.5.5 Data Exploration
4.5.6 Tasks
4.5.7 Preliminary Steps in Preparation for PCA
4.5.8 Complete PCA
4.5.9 Results Visualization
4.5.10 Results Interpretation
4.5.11 Conclusion
5 Future Research Directions
6 Conclusion
Key Terms and Definitions
References
Forecasting Returns of Crypto Currency: Identifying Robustness of Auto Regressive and Integrated Moving Average (ARIMA) and Ar...
1 Background and Motivation of Study
1.1 Historical Background of Bitcoin
2 Methods and Models Used for Analysis
2.1 Descriptive Statistics and Boxplots
2.2 Random Walk Model: Augmented Dickey-Fuller Test (ADF Test)
2.3 Auto Regressive Integrated Moving Average (ARIMA)
2.4 Artificial Neural Networks (ANNs)
3 Empirical Analysis
3.1 Forecasting of Bitcoin Through ARIMA Model
3.2 Forecasting of Bitcoin Through ANNs
3.3 Robustness Models Applied
4 Conclusion
References
Report
Part II: Machine Learning
Machine Learning in Financial Markets: Dimension Reduction and Support Vector Machine
1 Introduction
2 Background
2.1 Support Vector Machine (SVM) And Kernel SVM
2.2 Lasso
2.3 PCA
2.4 Kernel PCA
2.5 Probabilistic PCA
2.6 Sliced Inverse Regression (SIR)
2.7 Multidimensional Scaling (MDS)
2.8 Laplacian Eigenmaps
2.9 Local Linear Embedding
3 Dimension Reduction in Financial Markets
3.1 Modern Portfolio Optimization
3.2 Scenario Generation for Portfolio Theory
3.3 Algorithmic Trading
3.4 Future Research Directions
4 Conclusion
References
Pruned Random Forests for Effective and Efficient Financial Data Analytics
1 Introduction
2 Machine Learning: An Overview
2.1 Machine Learning Algorithms
2.2 Random Forest
2.3 Machine Learning Applications
3 Machine Learning in Financial Data Analytics
3.1 Credit Risk Management
3.2 Financial Fraud Detection
3.3 Portfolio Management
3.4 Real Estate Valuation
3.5 Insurance
3.6 Retail Banking
4 Pruned Random Forests
4.1 Clustering-Based Diverse Random Forest
4.1.1 Best Representative on Training CLUB-DRF
4.1.2 Best Representative on OOB CLUB-DRF
4.1.3 Random Representative CLUB-DRF
4.2 eGAP
5 Experimental Study
5.1 Datasets
5.2 Classification Datasets
5.3 Regression Datasets
5.4 Space and Inference Time Performance
6 Conclusion and Future Work
References
Foreign Currency Exchange Rate Prediction Using Long Short-Term Memory, Support Vector Regression and Random Forest Regression
1 Introduction
2 Related Works
3 Methodology
3.1 Dataset
3.2 Performance Measures
3.3 Forecasting Algorithm
3.3.1 Support Vector Machine
3.3.2 Random Forest
3.3.3 LSTM
4 Result and Discussion
5 Discussion
6 Conclusion and Future Works
References
Natural Language Processing for Exploring Culture in Finance: Theory and Applications
1 Introduction
2 Background
3 NLP Algorithms for Studying Culture in Finance
3.1 Basics
3.2 Sentiment Analysis
3.3 TF-IDF
3.4 Cosine Similarity
3.5 Word Embeddings
3.6 LDA Topic Models
4 Solutions and Recommendations
4.1 Overview
4.2 Semantic Analysis
5 Future Research Directions
6 Conclusion
References
Part III: Technology-Driven Finance
Network Modeling: Historical Perspectives, Agent-Based Modeling, Correlation Networks, and Network Similarities
1 Introduction
2 Financial Networks Literature: Historical Overview
3 Agent-Based Modeling in Economics and Finance
4 Application: Applying Schelling Model to Renting Decisions
5 Correlation-Based Stock Networks
6 Application: Topological Properties of Correlation Networks
7 Network Similarities Literature
7.1 Known Node-Correspondence (KNC) Methods
7.2 Unknown Node-Correspondence (UNC) Methods
8 Application: Network Similarities of Correlation-Based Stock Networks
9 Conclusion
References
Optimization of Regulatory Economic-Capital Structured Portfolios: Modeling Algorithms, Financial Data Analytics, and Reinforc...
1 Introduction
2 Review of Theoretical Foundations and Modeling Parameters Using Al Janabi Model
3 Scenario Optimization Framework of Structured Regulatory Economic-Capital Portfolios
3.1 Nonlinear and Dynamic Optimization of Regulatory Economic-Capital Optimal & Investable Portfolios Using LVaR Modeling Algo...
3.1.1 Variable Dynamics and Constraints for Scenario Optimization
3.1.2 Empirical Constrained Scenario Optimization of Regulatory Economic-Capital: The Case of Long and Short-Sales Structured ...
4 Concluding Remarks
References
Transforming Insurance Business with Data Science
1 Introduction
2 Data Science Challenges
2.1 Explore Analytics Opportunities
2.1.1 Find the Diamond
2.1.2 Leverage New Data
2.1.3 Predict the Next Move
2.1.4 Reduce Uncertainty
2.2 Evaluate Analytics Solutions
2.2.1 Operational Feasibility
2.2.2 Financial Viability
2.3 Prioritize Analytics Projects
2.3.1 Evaluation Framework
2.3.2 Project Selection Heuristic
2.3.3 Project Sequencing
2.4 Manage an Analytics Project
2.4.1 Scope and Hypotheses
2.4.2 Data Collection
2.4.3 Feature Engineering
2.4.4 Univariate
2.4.5 Multivariate Model Development
2.4.6 Model Evaluation
2.4.7 Model Deployment
2.4.8 Measure Business Impact
3 Build a Data Science Team
3.1 Data Science Functions
3.2 Data Science Organization
4 Analytics Applications
5 Ethical Considerations
6 Summary
References
A General Cyber Hygiene Approach for Financial Analytical Environment
1 Introduction
2 Background
3 Proposed Methodology
3.1 Cloud Computing Services
3.2 Current Services for Cloud Services
3.3 Infrastructure as a Service
3.4 Platform as a Service
3.5 Software as a Service
3.6 Open Source-Based Services
3.7 Main Characteristics of Cloud Computing Services
3.8 Common Characteristics
3.9 License Type
3.10 Intended User Group
3.11 Security and Privacy
3.12 Payment Systems
3.13 Specific Characteristics
3.14 IaaS-Specific Characteristics
3.15 PaaS-Specific Characteristics
3.16 SaaS-Specific Characteristics
4 Conclusion
References