Course 2: Advanced Data Science

Advanced Data Science provides an in-depth discussion of machine learning/artificial intelligence models and their applications to quantitative finance.

In particular, this course covers the following topics:

Machine learning models for point and probabilistic prediction, including supervised regression and classification, unsupervised autoencoders and graphical models
Decision theory and estimation risk assessment via cross-validation and Monte Carlo: variance /bias tradeoffs
Feature engineering and enhancements: feature bases, trees, neural networks, gradient boosting, lasso/ridge regularization, random forests, etc.
Financial applications of machine learning models for hedging, credit default and clustering

After the course you will obtain a Statement of Completion. Furthermore, you can choose to complete an optional code-based project that will count as one of the two Practical Projects towards the ARPM Certification.

Decision theory

• Statistical decision problems

• Bayes risk

• Minimax and other approaches

Machine learning foundations

• Foundations

• Approaches

• Prediction

• Learning and inference

Estimation assessment and enhancements

• Estimators as decisions

• Estimation assessment

• Bias versus variance

• Regularization

• Ensemble learning

Supervised learning - regression

• Supervised learning: regression

• Point least squares regression

• Probabilistic regression

Supervised learning - classification

• Supervised learning: classification

• Point binary classification

• Probabilistic classification

Unsupervised learning

• Least squares autoencoders

• Graphical models: introduction

• Graphical models: probabilistic factor analysis

• Graphical models: mixture models

• Graphical models: naive Bayes models

• Graphical models: Bayes networks

• Graphical models: Markov networks