Data Science for Economics and Finance: Methodologies and Applications

✍ Scribed by Sergio Consoli (editor), Diego Reforgiato Recupero (editor), Michaela Saisana (editor)

• Publisher: Springer
• Year: 2021
• Tongue: English
• Leaves: 357
• Edition: 1st ed. 2021
• Category: Library

✦ Synopsis

This open access book covers the use of data science, including advanced machine learning, big data analytics, Semantic Web technologies, natural language processing, social media analysis, time series analysis, among others, for applications in economics and finance. In addition, it shows some successful applications of advanced data science solutions used to extract new knowledge from data in order to improve economic forecasting models.

The book starts with an introduction on the use of data science technologies in economics and finance and is followed by thirteen chapters showing success stories of the application of specific data science methodologies, touching on particular topics related to novel big data sources and technologies for economic analysis (e.g. social media and news); big data models leveraging on supervised/unsupervised (deep) machine learning; natural language processing to build economic and financial indicators; and forecasting and nowcasting of economic variables through time series analysis.  

This book is relevant to all stakeholders involved in digital and data-intensive research in economics and finance, helping them to understand the main opportunities and challenges, become familiar with the latest methodological findings, and learn how to use and evaluate the performances of novel tools and frameworks. It primarily targets data scientists and business analysts exploiting data science technologies, and it will also be a useful resource to research students in disciplines and courses related to these topics. Overall, readers will learn modern and effective data science solutions to create tangible innovations for economic and financial applications.

✦ Table of Contents

Foreword
Preface
How This Book Is Organized
Target Audience
Acknowledgments
Contents
Data Science Technologies in Economics and Finance: A Gentle Walk-In
1 Introduction
2 Technical Challenges
2.1 Stewardship and Protection
2.2 Data Quantity and Ground Truth
2.3 Data Quality and Provenance
2.4 Data Integration and Sharing
2.5 Data Management and Infrastructures
3 Data Analytics Methods
3.1 Deep Machine Learning
3.2 Semantic Web Technologies
4 Conclusions
References
Supervised Learning for the Prediction of Firm Dynamics
1 Introduction
2 Supervised Machine Learning
3 SL Prediction of Firm Dynamics
3.1 Entrepreneurship and Innovation
3.2 Firm Performance and Growth
3.3 Financial Distress and Firm Bankruptcy
4 Final Discussion
References
Opening the Black Box: Machine Learning Interpretability and Inference Tools with an Application to Economic Forecasting
1 Introduction
2 Data and Experimental Setup
2.1 Data
2.2 Models
2.3 Experimental Procedure
3 Forecasting Performance
3.1 Baseline Setting
3.2 Robustness Checks
4 Model Interpretability
4.1 Methodology
4.1.1 Permutation Importance
4.1.2 Shapley Values and Regressions
4.2 Results
4.2.1 Feature Importance
4.2.2 Shapley Regressions
5 Conclusion
References
Machine Learning for Financial Stability
1 Introduction
2 Overview of Machine Learning Approaches
3 Tree Ensembles
3.1 Decision Trees
3.2 Random Forest
3.3 Tree Boosting
3.4 CRAGGING
4 Regularization, Shrinkage, and Sparsity
4.1 Regularization
4.2 Bayesian Learning
5 Critical Discussion on Machine Learning as a Tool for Financial Stability Policy
6 Literature Overview
6.1 Decision Trees for Financial Stability
6.2 Sparse Models for Financial Stability
6.3 Unsupervised Learning for Financial Stability
7 Conclusions
References
Sharpening the Accuracy of Credit Scoring Models with Machine Learning Algorithms
1 Introduction
2 Preliminaries and Linear Methods for Classification
2.1 Logistic Regression
2.2 Linear Discriminant Analysis
2.3 Naïve Bayes
3 Nonlinear Methods for Classification
3.1 Decision Trees
3.2 Neural Networks
3.3 Support Vector Machines
3.4 k-Nearest Neighbor
3.5 Genetic Algorithms
3.6 Ensemble Methods
4 Comparison of Classifiers in Credit Scoring Applications
4.1 Comparison of Individual Classifiers
4.2 Comparison of Ensemble Classifiers
4.3 One-Class Classification Methods
5 Conclusion
References
Classifying Counterparty Sector in EMIR Data
1 Introduction
2 Reporting Under EMIR
3 Methodology
3.1 First Step: The Selection of Data Sources
3.2 Second Step: Data Harmonisation
3.3 Third Step: The Classification
3.3.1 Classifying Commercial and Investment Banks
3.3.2 Classifying Investment Funds
3.4 Description of the Algorithm
4 Results
5 Applications
5.1 Case Study I: Use of Derivatives by EA Investment Funds
5.2 Case Study II: The Role of Commercial and Investment Banks
5.3 Case Study III: The Role of G16 Dealers in the EA Sovereign CDS Market
5.4 Case Study IV: The Use of Derivatives by EA Insurance Companies
References
Massive Data Analytics for Macroeconomic Nowcasting
1 Introduction
2 Review of the Recent Literature
2.1 Various Types of Massive Data
2.2 Econometric Methods to Deal with Massive Datasets
3 Example of Macroeconomic Applications Using Massive Alternative Data
3.1 A Real-Time Proxy for Exports and Imports
3.1.1 International Trade
3.1.2 Localization Data
3.1.3 QuantCube International Trade Index: The Case of China
3.2 A Real-Time Proxy for Consumption
3.2.1 Private Consumption
3.2.2 Alternative Data Sources
3.2.3 QuantCube Chinese Tourism Index
3.3 A Real-Time Proxy for Activity Level
3.3.1 Satellite Images
3.3.2 Pre-processing and Modeling
3.3.3 QuantCube Activity Level Index
4 High-Frequency GDP Nowcasting
4.1 Nowcasting US GDP
4.2 Nowcasting Chinese GDP
5 Applications in Finance
6 Conclusions
References
New Data Sources for Central Banks
1 Introduction
2 New Data Sources for Central Banks
3 Successful Case Studies
3.1 Newspaper Data: Measuring Uncertainty
3.1.1 Economic Policy Uncertainty in Spain
3.1.2 Economic Policy Uncertainty in Latin America
3.2 The Narrative About the Economy as a Shadow Forecast: An Analysis Using the Bank of Spain Quarterly Reports
3.3 Forecasting with New Data Sources
3.3.1 A Supervised Method
3.3.2 An Unsupervised Method
3.3.3 Google Forecast Trends of Private Consumption
4 Conclusions
References
Sentiment Analysis of Financial News: Mechanics and Statistics
1 Introduction
1.1 Brief Background on Sentiment Analysis in Finance
2 Mechanics of Textual Sentiment Analysis
3 Statistics of Sentiment Indicators
3.1 Stylized Facts
3.2 Statistical Tests and Models
3.2.1 Independence
3.2.2 Stationarity
3.2.3 Causality
3.2.4 Variable Selection
4 Empirical Analysis
5 Software
References
Semi-supervised Text Mining for Monitoring the News About the ESG Performance of Companies
1 Introduction
2 Methodology to Create Text-Based Indicators
2.1 From Text to Numerical Data
2.1.1 Keywords Generation
2.1.2 Database Querying
2.1.3 News Filtering
2.1.4 Indicators Construction
2.2 Validation and Decision Making
3 Monitoring the News About Company ESG Performance
3.1 Motivation and Applications
3.1.1 Text-Based ESG Scoring as a Risk Management Tool
3.1.2 Text-Based ESG Scoring as an Investment Tool
3.2 Pipeline Tailored to the Creation of News-Based ESG Indices
3.2.1 Word Embeddings and Keywords Definition
3.2.2 Company Selection and Corpus Creation
3.2.3 Aggregation into Indices
3.2.4 Validation
3.3 Stock and Sector Screening
3.3.1 Aggregate Portfolio Performance Analysis
3.3.2 Additional Analysis
4 Conclusion
References
Extraction and Representation of Financial Entities from Text
1 Introduction
2 Extracting Knowledge Graphs from Text
2.1 Named Entity Recognition (NER)
2.2 Named Entity Linking (NEL)
2.3 Relationship Extraction (RELEX)
3 Refining the Knowledge Graph
4 Analyzing the Knowledge Graph
5 Exploring the Knowledge Graph
6 Semantic Exploration Using Visualizations
7 Conclusion
References
Quantifying News Narratives to Predict Movements in Market Risk
1 Introduction
2 Preliminaries
2.1 Topic Modeling
2.1.1 Latent Dirichlet Analysis
2.1.2 Paragraph Vector
2.1.3 Gaussian Mixture Models
2.2 Gradient Boosted Trees
2.3 Market Risk and the CBOE Volatility Index (VIX)
3 Methodology
3.1 News Data Acquisition and Preparation
3.2 Narrative Extraction and Topic Modeling
3.2.1 Approach 1: Narrative Extraction Using Latent Dirichlet Analysis
3.2.2 Approach 2: Narrative Extraction Using Vector Embedding and Gaussian Mixture Models
3.3 Predicting Movements in Market Risk with Machine Learning
3.4 Evaluation on Time Series
4 Experimental Results and Discussion
4.1 Feature Setups and Predictive Performance
4.2 The Effect of Different Prediction Horizons
5 Conclusion
References
Do the Hype of the Benefits from Using New Data Science Tools Extend to Forecasting Extremely Volatile Assets?
1 Introduction
2 What Is Bitcoin?
3 Bitcoin Data and HAR-Type Strategies to Forecast Volatility
4 Machine Learning Strategy to Forecast Volatility
5 Social Media Data
6 Empirical Exercise
7 Robustness Check
7.1 Different Window Lengths
7.2 Different Sample Periods
7.3 Different Tuning Parameters
7.4 Incorporating Mainstream Assets as Extra Covariates
8 Conclusion
Appendix: Data Resampling Techniques
References
Network Analysis for Economics and Finance: An application to Firm Ownership
1 Introduction
2 Network Analysis in the Literature
3 Network Analysis
4 Network Analysis: An Application to Firm Ownership
4.1 Data
4.2 Network Construction
4.3 Network Statistics
4.4 Bow-Tie Structure
5 Conclusion
Appendix
References