Παρουσίαση/Προβολή

Εικόνα επιλογής

MSc in Statistics AUEB Short Courses: Machine Learning in R: Applications in Finance

(SC1153) -  Ionut Florescu/Ioannis Ntzoufras

Περιγραφή Μαθήματος

This eclass is created for the Short course of the M.Sc. in Statistics of the Department of Statistics @ AUEB  

 

Current Course November 2022: Machine Learning in R: Applications in Finance 

by 

Ionut Florescu,

Research Professor, Director of the M.Sc. in Financial Analytics

Stevens Institute of Technology (USA)

 

Registration is required for all non-M.Sc. in Stats participants

 

Description of Lectures:

Lecture 1: What is Corporate Credit Rating? Comparing basic Machine Learning techniques.

In this lecture we will discuss how machine learning techniques may be used to assess corporate credit ratings. We will discuss details of the article Golbayani, Parisa, Ionut¸ Florescu, and Rupak Chatterjee (2020). “A comparative study of forecasting corporate credit ratings using neural networks, support vector machines, and decision trees”. In: The North American Journal of Economics and Finance 54, p. 101251.

In the practical session we will introduce the data used throughout the lectures and learn how the Support Vector Machine algorithm may be used in this area.

 

Lecture 2: Corporate Credit Rating. Comparing more advanced Machine Learning Techniques.

In this lecture we will discuss deep network architectures specifically LSTM and CNN. We will discuss details of the article Golbayani, Parisa, Dan Wang, and Ionut Florescu (2021). “Application of deep neural networks to assess corporate credit rating”. In:

International Journal of Mechanical and Industrial Engineering 14(1). url: https://arxiv.org/abs/2003.02334.

In the practical section we will discuss the most basic Artificial Neural Network architecture (Multi Layer Perceptron) and apply it to financial data.

 

Lecture 3: Corporate Credit Learning. Understanding how inputs change the output: Counterfactual Explanation.

In this lecture we will discuss the results obtained in our most recent research Wang, Dan, Zhi Chen, and Ionut Florescu (2021). A Sparsity Algorithm with Applications to Corporate Credit Rating. url: https://arxiv.org/abs/2107.10306.

We will present an algorithm that may be used to determine the smallest modification to input variables to change the classification given an existing ML algorithm.

In the practical session we will discuss Decision trees particularly Random forest. We will apply RF to financial data.

 

All participants should bring their laptops in all three lectures and have R and Rstudio installed.

  1. Install R from the University of Crete mirror: https://ftp.cc.uoc.gr/mirrors/CRAN/ (only base)
  2. Download and install Rstudio Desktop https://posit.co/download/rstudio-desktop/

 

Ημερομηνία δημιουργίας

Παρασκευή 11 Νοεμβρίου 2022