Data Science and introduction to R

Course objectives

Presentation of basic methods in the area of statistical data analysis and machine learning, along with typical applications in the area of credit risk.

Benefits

Participants of the training will learn:​

• what kinds of statistical and AI/ML methods are used in banking, especially for credit risk analysis​
• how to use the R language​
• how to create data science and machine learning models in R​
• how to use R in the area of credit risk

Training scope

• Introduction to statistical data analysis
  • Introduction to Data Science, Machine Learning and Artificial Intelligence​
  • Development of DS/ML/AI in the last 25 years​
  • Key statistical concepts​
  • Types of statistical models​
  • Linear and nonlinear models​
  • Estimation methods​
  • Methods of validation and model selection​
  • Principles of statistical tests​
• Introduction to Machine Learning
  • Regression, classification, learning​
  • Discrete choice models​
  • Unbalanced sampling​
  • Decision trees, random forests, SVMs​
  • Empirical and structural risk (SLT)​
  • Regularization​
  • AI and machine learning​
  • Artificial neural networks​
  • Types of neural networks​
  • Neural network learning methods​
  • Principles of deep learning​
  • Modern neural network architectures​
  • Generative models​
  • LLM models
• Introduction to programming with R
  • R and R Studio​
  • Loading data​
  • Data structures and functions in R​
  • Data processing​
  • Regression models, decision trees, random forests, SVMs​
  • Artificial neural networks – construction and learning​
  • Deep learning – construction and learning​
  • Combining R with AI learning environments

The work in R will be based on specially prepared case studies from the area of credit risk.

Audience

Analysts and those working with data analytics and machine learning specialists looking to gain an understanding of the fundamentals of such methods, with an emphasis on practical applications in the area of credit risk.