Credit Scoring - Business Process Automation

The combination of statistics (descriptive and logistic regression), AI/ML (Artificial Intelligence and Machine Learning: elements of decision trees, XGBoosting, random forests, deep learning and variable selection heuristics) with programming in SAS and OpenSource is presented. During the course, software in SAS and OpenSource for building predictive models and simulating the credit approval process is improved and modified. Each class is usually a presentation of a program and algorithm along with a business interpretation.

The primary goal of the course is to provide students with experience and skills in automating the process in which credit decisions are automatically made based on a predictive model. It is therefore about building the Data Driven Decision Making environment. Therefore, all the most important technical, analytical and business aspects of the entire process are presented here. After completing this course, it can be expected that the student will fully control the entire business process, from its business and financial assumptions (also calculating the financial result) and ending with the implementation of the model and verification of assumptions with the final effect. It presents the construction of a risk scorecard and PD (default probability) models, study of programs and entering into all the details of each of the construction elements. The goal, therefore, appears as a presentation of a method for self-development of tools in order to be able to fully understand and control each stage of building models and not be dependent on specific software. Another aim is also to present any business problems related to the credit approval/acceptance process and the complex process of acquisition and cross-selling. During the course, the student learns to analyze the acceptance simulation process and can experience the consequences of changes in the parameters of the process. Thanks to the special simulated data used during classes, the student can feel like the Director of the Credit Risk Department and manage a banking process in which he either brings profits to the bank or loses them. The final project is a type of strategy game where each project team tries to define the parameters of the acceptance process in such a way as to maximize profits and earn more than the initial model.

Literatura podstawowa:

Credit scoring in the context of interpretable machine learning. Theory and practice. Edited by D. Kaszyński, B. Kamiński, T. Szapiro. Oficyna Wydawnicza SGH, Warszawa 2020 (https://ssl-kolegia.sgh.waw.pl/pl/KAE/struktura/IE/struktura/ZWiAD/publikacje/Documents/Credit_scoring_in_the_context_of_interpretable_machine_learning.pdf).

K. Przanowski, Credit Scoring - Studia przypadków procesów biznesowych, 2015, SGH - in Polish

K. Przanowski, Credit Scoring w erze Big Data, OW SGH, 2014 - in Polish

N. Siddiqi, Credit risk scorecards: Developing and implementing intelligent credit scoring. Wiley and SAS Business Series, 2005;

L.C. Thomas, D.B. Edelman, J.N. Crook, Credit Scoring and Its Applications, Society for Industrial and Applied Mathematics, Philadelfia 2002;

Basel Committee on Banking Supervision, Working paper no. 14, 2005; Studies on the validation of internal rating systems, Bank for International Settlements.

Literatura uzupełniająca:

Papers from the conference CRC (Credit Research Center): http://www.business-school.ed.ac.uk/crc/conferences/

A. Matuszyk, Credit Scoring, SGH, Warszawa 2009; E. Frątczak, red., 2012, Zaawansowane metody analiz statystycznych, SGH,

SAS Online Doc, SAS Institute Inc., http://support.sas.com/onlinedoc/913/docMainpage.jsp

Kryteria oceniania:

egzamin testowy: 50.00%

projekty: 50.00%

The combination of statistics (descriptive and logistic regression), AI/ML (Artificial Intelligence and Machine Learning: elements of decision trees, XGBoosting, random forests, deep learning and variable selection heuristics) with programming in SAS and OpenSource is presented. During the course, software in SAS and OpenSource for building predictive models and simulating the credit approval process is improved and modified. Each class is usually a presentation of a program and algorithm along with a business interpretation.

The primary goal of the course is to provide students with experience and skills in automating the process in which credit decisions are automatically made based on a predictive model. It is therefore about building the Data Driven Decision Making environment. Therefore, all the most important technical, analytical and business aspects of the entire process are presented here. After completing this course, it can be expected that the student will fully control the entire business process, from its business and financial assumptions (also calculating the financial result) and ending with the implementation of the model and verification of assumptions with the final effect. It presents the construction of a risk scorecard and PD (default probability) models, study of programs and entering into all the details of each of the construction elements. The goal, therefore, appears as a presentation of a method for self-development of tools in order to be able to fully understand and control each stage of building models and not be dependent on specific software. Another aim is also to present any business problems related to the credit approval/acceptance process and the complex process of acquisition and cross-selling. During the course, the student learns to analyze the acceptance simulation process and can experience the consequences of changes in the parameters of the process. Thanks to the special simulated data used during classes, the student can feel like the Director of the Credit Risk Department and manage a banking process in which he either brings profits to the bank or loses them. The final project is a type of strategy game where each project team tries to define the parameters of the acceptance process in such a way as to maximize profits and earn more than the initial model.

Literatura podstawowa:

Credit scoring in the context of interpretable machine learning. Theory and practice. Edited by D. Kaszyński, B. Kamiński, T. Szapiro. Oficyna Wydawnicza SGH, Warszawa 2020 (https://ssl-kolegia.sgh.waw.pl/pl/KAE/struktura/IE/struktura/ZWiAD/publikacje/Documents/Credit_scoring_in_the_context_of_interpretable_machine_learning.pdf).

K. Przanowski, Credit Scoring - Studia przypadków procesów biznesowych, 2015, SGH - in Polish

K. Przanowski, Credit Scoring w erze Big Data, OW SGH, 2014 - in Polish

N. Siddiqi, Credit risk scorecards: Developing and implementing intelligent credit scoring. Wiley and SAS Business Series, 2005;

L.C. Thomas, D.B. Edelman, J.N. Crook, Credit Scoring and Its Applications, Society for Industrial and Applied Mathematics, Philadelfia 2002;

Basel Committee on Banking Supervision, Working paper no. 14, 2005; Studies on the validation of internal rating systems, Bank for International Settlements.

Literatura uzupełniająca:

Papers from the conference CRC (Credit Research Center): http://www.business-school.ed.ac.uk/crc/conferences/

A. Matuszyk, Credit Scoring, SGH, Warszawa 2009; E. Frątczak, red., 2012, Zaawansowane metody analiz statystycznych, SGH,

SAS Online Doc, SAS Institute Inc., http://support.sas.com/onlinedoc/913/docMainpage.jsp

Kryteria oceniania:

egzamin testowy: 50.00%

projekty: 50.00%

The combination of statistics (descriptive and logistic regression), AI/ML (Artificial Intelligence and Machine Learning: elements of decision trees, XGBoosting, random forests, deep learning and variable selection heuristics) with programming in SAS and OpenSource is presented. During the course, software in SAS and OpenSource for building predictive models and simulating the credit approval process is improved and modified. Each class is usually a presentation of a program and algorithm along with a business interpretation.

The primary goal of the course is to provide students with experience and skills in automating the process in which credit decisions are automatically made based on a predictive model. It is therefore about building the Data Driven Decision Making environment. Therefore, all the most important technical, analytical and business aspects of the entire process are presented here. After completing this course, it can be expected that the student will fully control the entire business process, from its business and financial assumptions (also calculating the financial result) and ending with the implementation of the model and verification of assumptions with the final effect. It presents the construction of a risk scorecard and PD (default probability) models, study of programs and entering into all the details of each of the construction elements. The goal, therefore, appears as a presentation of a method for self-development of tools in order to be able to fully understand and control each stage of building models and not be dependent on specific software. Another aim is also to present any business problems related to the credit approval/acceptance process and the complex process of acquisition and cross-selling. During the course, the student learns to analyze the acceptance simulation process and can experience the consequences of changes in the parameters of the process. Thanks to the special simulated data used during classes, the student can feel like the Director of the Credit Risk Department and manage a banking process in which he either brings profits to the bank or loses them. The final project is a type of strategy game where each project team tries to define the parameters of the acceptance process in such a way as to maximize profits and earn more than the initial model.

Literatura podstawowa:

Credit scoring in the context of interpretable machine learning. Theory and practice. Edited by D. Kaszyński, B. Kamiński, T. Szapiro. Oficyna Wydawnicza SGH, Warszawa 2020 (https://ssl-kolegia.sgh.waw.pl/pl/KAE/struktura/IE/struktura/ZWiAD/publikacje/Documents/Credit_scoring_in_the_context_of_interpretable_machine_learning.pdf).

K. Przanowski, Credit Scoring - Studia przypadków procesów biznesowych, 2015, SGH - in Polish

K. Przanowski, Credit Scoring w erze Big Data, OW SGH, 2014 - in Polish

N. Siddiqi, Credit risk scorecards: Developing and implementing intelligent credit scoring. Wiley and SAS Business Series, 2005;

L.C. Thomas, D.B. Edelman, J.N. Crook, Credit Scoring and Its Applications, Society for Industrial and Applied Mathematics, Philadelfia 2002;

Basel Committee on Banking Supervision, Working paper no. 14, 2005; Studies on the validation of internal rating systems, Bank for International Settlements.

Literatura uzupełniająca:

Papers from the conference CRC (Credit Research Center): http://www.business-school.ed.ac.uk/crc/conferences/

A. Matuszyk, Credit Scoring, SGH, Warszawa 2009; E. Frątczak, red., 2012, Zaawansowane metody analiz statystycznych, SGH,

SAS Online Doc, SAS Institute Inc., http://support.sas.com/onlinedoc/913/docMainpage.jsp

Kryteria oceniania:

egzamin testowy: 50.00%

projekty: 50.00%

The combination of statistics (descriptive and logistic regression), AI/ML (Artificial Intelligence and Machine Learning: elements of decision trees, XGBoosting, random forests, deep learning and variable selection heuristics) with programming in SAS and OpenSource is presented. During the course, software in SAS and OpenSource for building predictive models and simulating the credit approval process is improved and modified. Each class is usually a presentation of a program and algorithm along with a business interpretation.

The primary goal of the course is to provide students with experience and skills in automating the process in which credit decisions are automatically made based on a predictive model. It is therefore about building the Data Driven Decision Making environment. Therefore, all the most important technical, analytical and business aspects of the entire process are presented here. After completing this course, it can be expected that the student will fully control the entire business process, from its business and financial assumptions (also calculating the financial result) and ending with the implementation of the model and verification of assumptions with the final effect. It presents the construction of a risk scorecard and PD (default probability) models, study of programs and entering into all the details of each of the construction elements. The goal, therefore, appears as a presentation of a method for self-development of tools in order to be able to fully understand and control each stage of building models and not be dependent on specific software. Another aim is also to present any business problems related to the credit approval/acceptance process and the complex process of acquisition and cross-selling. During the course, the student learns to analyze the acceptance simulation process and can experience the consequences of changes in the parameters of the process. Thanks to the special simulated data used during classes, the student can feel like the Director of the Credit Risk Department and manage a banking process in which he either brings profits to the bank or loses them. The final project is a type of strategy game where each project team tries to define the parameters of the acceptance process in such a way as to maximize profits and earn more than the initial model.

Literatura podstawowa:

Credit scoring in the context of interpretable machine learning. Theory and practice. Edited by D. Kaszyński, B. Kamiński, T. Szapiro. Oficyna Wydawnicza SGH, Warszawa 2020 (https://ssl-kolegia.sgh.waw.pl/pl/KAE/struktura/IE/struktura/ZWiAD/publikacje/Documents/Credit_scoring_in_the_context_of_interpretable_machine_learning.pdf).

K. Przanowski, Credit Scoring - Studia przypadków procesów biznesowych, 2015, SGH - in Polish

K. Przanowski, Credit Scoring w erze Big Data, OW SGH, 2014 - in Polish

N. Siddiqi, Credit risk scorecards: Developing and implementing intelligent credit scoring. Wiley and SAS Business Series, 2005;

L.C. Thomas, D.B. Edelman, J.N. Crook, Credit Scoring and Its Applications, Society for Industrial and Applied Mathematics, Philadelfia 2002;

Basel Committee on Banking Supervision, Working paper no. 14, 2005; Studies on the validation of internal rating systems, Bank for International Settlements.

Literatura uzupełniająca:

Papers from the conference CRC (Credit Research Center): http://www.business-school.ed.ac.uk/crc/conferences/

A. Matuszyk, Credit Scoring, SGH, Warszawa 2009; E. Frątczak, red., 2012, Zaawansowane metody analiz statystycznych, SGH,

SAS Online Doc, SAS Institute Inc., http://support.sas.com/onlinedoc/913/docMainpage.jsp

Kryteria oceniania:

egzamin testowy: 50.00%

projekty: 50.00%