This book is an introduction to machine learning using Python programming language with applications in finance and business. Coverages include the prediction methods of logistic regression, Naïve Bayes, k-Nearest Neighbor, Support Vector Machine, Random Forest, Gradient Boosting, and various types of Neural Networks. Performance measurements and assessments of feature importance are also explained. The book also contains detailed examples of the applications with data. Python codes are explained in a step-by-step manner using Jupyter Notebook so that the readers can practise on their own.
Contents:
- Financial Data Correlations:
- Portfolio Diversification
- Worked Example: Data
- Forming Optimal Portfolios
- Denoising the Correlation Matrix
- Using a More Accurate Forward Predictor
- Concluding Thoughts
- Regression and Regularization:
- Regression and Regularization
- Worked Example: Data
- Linear Regression Prediction
- Fine-Tuning Hyperparameters
- Prediction Using Hold-Out Test Set
- Cross-Validation
- Concluding Thoughts
- Corporate Reporting Data and Analyses:
- Financial Statements
- Binary Classification Performance Metrics
- Logistic Regression
- Worked Example: Data
- Dimension Reduction and Principal Component Analysis
- Concluding Thoughts
- Naïve Bayes, k-NN, and Support Vector Machines:
- Naïve Bayes
- k-Nearest Neighbors Algorithm
- Support Vector Machine
- Lagrange Duality Solution
- Support Vector Regression
- Worked Examples: Data
- Concluding Thoughts
- Decision Trees, Random Forest, and Multi-Class Prediction:
- Decision Tree Method
- Worked Example DT: Data
- Random Forest
- Multi-Class Classification
- Concluding Thoughts
- Gradient Boosting, SHAP Values:
- Gradient Boosting
- Popular Alternatives
- Worked Example GB: Data
- Shapley and SHAP Values
- Concluding Thoughts
- Artificial Neural Network I:
- Forward Propagation
- Backward Propagation
- Worked Example: Data
- Improving the NN and Understanding the Res
- Concluding Thoughts
- Artificial Neural Network II:
- Recurrent Neural Network
- Worked Example I: Data
- Variants of RNN
- Worked Example II: Data
- Concluding Thoughts
Readership: For undergraduate and graduate students in Machine Learning and Algorithms, Quantitative Finance, Computational Finance, Machine Learning, and Business Finance, as well as general public readers who want to improve their general knowledge on Machine Learning.
