Building Scoring Models Using Machine Learning Methods

2023/2024

Type: Mago-Lego

Delivered by: Department of Business Informatics

When: 2, 3 module

Open to: students of one campus

Instructors: Ilya Munerman

Language: English

ECTS credits: 6

Contact hours: 48

Full Syllabus

Abstract

The main aim of this course is a species of different country data sets and technics for integration this data for common international data environment. It also considers the features of algorithms for use online data for the instant response of the model to changing environmental circumstances, taken into the consequences of a pandemic. The course is designed for listeners which known elementary economics, finances, IT and mathematics and may be able for economists, IT specialists, managers, include MBA and journalists.

Learning Objectives

Ability to use modern methods and relevant information technologies to formulate and solve fintech problems
Master data-driven services marketing skills
Ability to estimate the performance of financial decisions based on modern models and computer programs
Knowledge of the types of modern scoring systems, methods of their design and data sources
Ability to use data mining and machine learning methods to solve applied problems
Ability to design business models based on intelligent IT services

Expected Learning Outcomes

Describe the main types of scoring ratings, etc. common features and differences between them
Formulate and correctly interpret definitions of concepts, terms and categories used in the development of scoring models
Formulate what problems scoring models solve
Be able to quickly recognize problems and find the scoring model needed to solve it
Demonstrate the ability to solve design and economic problems in professional activity
Formulate the differences between due diligence, financial risk indices and comprehensive credit scoring
Be able to form a scoring line for solving a specific task of the subject of economic activity
Apply basic machine learning tools
Solve problems of building machine learning models using modern software
Apply main approaches to building scoring systems based on modern financial concepts, such as methods of residual income, claim burden, payment discipline indices, etc.
Calculate corrections to scoring models depending on the purposes of their construction, calculate credit limits, adjust the probability of occurrence of various events
Master the methods of Big Data analysis used to solve professional tasks at the micro, meso and macro levels, including at the level of the financial market
Apply algorithms of recommendation systems and black box interpreters
Formulate basic methods of scoring work in a large company or marketplace
Be able to organize comprehensive counterparty scoring, including monitoring of the dealer network and potential customers, develop recommendation systems and calculate the functional value of complex assets
Demonstrate the mastery of Fintech tools

Course Contents

Contemporary financial analysis main challenges
Big data, data mining, data science
Main data types
Correlation
Data processing. Modeling
Estimation and model testing
Interpretable models and recommendation systems
Applications

Assessment Elements

Home assignments
Individual data processing tasks
Case disscussions
When preparing for discussions in practical classes, it is necessary to use not only lecture material, educational literature, but also regulatory legal acts and materials of law enforcement practice. Theoretical material should be correlated with legal norms, since changes and additions may be made to them, which are not always reflected in the educational literature
Final exam
Test based on course materials
Kaggle contests
During the course, three closed championships are held in Kaggle, on the construction of scoring models of the value of real estate, the probability of default and another one in agreement with the level of students

Interim Assessment

2023/2024 3rd module
0.2 * Case disscussions + 0.3 * Final exam + 0.25 * Home assignments + 0.25 * Kaggle contests

Bibliography

Recommended Core Bibliography

Bernd Engelmann, & Ha Pham. (2020). Measuring the Performance of Bank Loans under Basel II/III and IFRS 9/CECL. Risks, 8(93), 93. https://doi.org/10.3390/risks8030093
Brooks,Chris. (2019). Introductory Econometrics for Finance. Cambridge University Press. Retrieved from http://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=edsrep&AN=edsrep.b.cup.cbooks.9781108422536
Ian Goodfellow and Yoshua Bengio and Aaron Courville. Deep Learning, 2016. URL: http://www.deeplearningbook.org
Siddiqi, N. (2017). Intelligent Credit Scoring : Building and Implementing Better Credit Risk Scorecards (Vol. 2nd edition). Hoboken, New Jersey: Wiley. Retrieved from http://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=edsebk&AN=1441143

Recommended Additional Bibliography

Gareth James, Daniela Witten, Trevor Hastie, & Robert Tibshirani. (2013). An Introduction to Statistical Learning : With Applications in R. Springer.
Greenwald, A., Nosek, B., & Banaji, M. (2016). Understanding and Using the Implicit Association Test: 1. An Improved Scoring Algorithm. Retrieved from http://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=edsbas&AN=edsbas.4AB789A7
Trevor Hastie, Robert Tibshirani , et al., The Elements of Statistical Learning: Data Mining, Inference, and Prediction, 2nd edition, 2017. Free from the publisher: https://web.stanford.edu/~hastie/ElemStatLearn/printings/ESLII_print12.pdf

Course Syllabus