SlideShare a Scribd company logo
1 of 48
Credit Risk with AI tools
The old, the new and the unexpected
ARMANDO VIEIRA
Armandosvieira.wordpress.com
Customer fails
to pay
Losing money
Wrong Strategy
Change in
market
prices
Processing failures and
frauds
Regulatory compliance
Customer fails
to pay
Losing money
Wrong Strategy
Change in
market
prices
Processing failures and
frauds
Regulatory compliance
RISK
Importance of Credit Risk
A statistical means of providing a quantifiable risk factor for a given
applicant.
Credit scoring is a process whereby information provided is converted into
numbers to arrive at a score.
The objective is to forecast future performance from past behavior of
clients (SME or individuals).
Credit scoring are used in many areas of industries:
Banking
Decision Models Finance
Insurance
Retail
Telecommunications
What is Credit Scoring?
ā€¢ Predict financial distress of private companies one year ahead
based on account balance sheet from previous years.
ā€¢ Enventualy the probability to become so.
ā€¢ Obtain reliable data from up to 5 previous years before failure
ā€¢ Classify and release warning signs
Bankruptcy prediction problem
The curse of dimensionality
Problems
ā€¢ Sparness of the search space
ā€¢ Presence of Irrelevant Features
ā€¢ Poor generalization of Learning Machine
ā€¢ Exceptions difficult to identify
Solutions
ā€¢ Dimensionality reduction: feature selection
ā€¢ Constrain the complexity of the Learning Machine
The Diane Database
ā€¢ Financial statements of French companies, initially of 60,000
industrial French companies, for the years of 2002 to 2006,
with at least 10 employees
ā€¢ 3,000 were declared bankrupted in 2007 or presented a
ā€¢ restructuring plan 30 financial ratios which allow the
description of firms in terms of the financial strength,
liquidity, solvability, productivity of labor and capital, margins,
net profitability and return on investment
The inputs
Number of employees Net Current Assets/Turnover (days)
Financial Debt / Capital Employed (%) Working Capital Needs / Turnover (%)
Capital Employed / Fixed Assets Export (%)
Depreciation of Tangible Assets (%) Value added per employee
Working capital / current assets Total Assets / Turnover
Current ratio Operating Profit Margin (%)
Liquidity ratio Net Profit Margin (%)
Stock Turnover days Added Value Margin (%)
Collection period Part of Employees (%)
Credit Period Return on Capital Employed (%)
Turnover per Employee Return on Total Assets (%)
Interest / Turnover EBIT Margin (%)
Debt Period (days) EBITDA Margin (%)
Financial Debt / Equity (%) Cashflow / Turnover (%)
Financial Debt / Cashflow Working Capital / Turnover (days)
Hard problem
0
2
4
6
3 4 5 6 7
Class 0
Class 1
Ī»
1
Ī»
2
First two principal component from PCA
How HLVQ-C works
0
0.5
1.0
1.5
0 0.5 1.0 1.5
Class 0
Class 1
After
Before
?
d2
d1
X
Y
DIANE 1 (error%)
Model Error I Error II Total
MDA
SVM
MLP
HLVQ-C
26.4
17.6
25.7
11.1
21.0
12.2
13.1
10.6
23.7
14.8
19.4
10.8
DIANE 1 - HLVQC Results
Method
Classification
Weighted Efficiency
(%)
Z-score (Altman) 62.7
Best Discriminant 66.1
MLP 71.4
OurMethod 84.1
Source: Vieira, A.S., Neves, J.C.: Improving Bankruptcy Prediction with Hidden Layer
Learning. Vector Quantization. European Accounting Review, 15 (2), 253-271 (2006).
Personal credit
Results I ā€“ 30 days into arrears
Classifier Accuracy (%) Type I Type II
G
Logistic 66.3 27.3 40.1
54.8
MLP 67.5 8.1 57.1
61.1
SVM 64.9 35.6 34.6
52.3
AdaboostM1 69.0 12.6 49.4
55.7
HLVQ-C 72.6 5.3 49.5
52.3
Results I ā€“ 60 days into arrears
Classifier Accuracy Type I Type II
G
Logistic 81.2 48.2 11.0
21.2
MLP 82.3 57.4 9.1
20.1
SVM 83.3 38.1 12.4
19.3
AdaboostM1 84.1 45.7 8.0
14.7
HLVQ-C 86.5 48.3 6.2
11.9
DIANE II (2002 ā€“ 2007)
ā€¢ More data
ā€¢ Longer history
ā€¢ More features
Year
2006
Classifier Accuracy Type I Type II
Logistic 91.25 6.33 11.17
MLP 91.17 6.33 11.33
C-SVM 92.42 5.16 10.00
AdaboostM1 89.75 8.16 12.33
Year
2005
Classifier Accuracy Type I Type II
Logistic 79.92 19.50 20.67
MLP 75.83 24.50 23.83
C-SVM 80.00 21.17 18.83
AdaboostM1 78.17 20.50 23.17
Results
How useful?
[ ]mexexNV III )1()1( āˆ’āˆ’āˆ’=Ī·
ļ£·ļ£·
ļ£ø
ļ£¶
ļ£¬ļ£¬
ļ£­
ļ£«
āˆ’
>>
āˆ’ I
II
e
e
mmG
x
x
11
The Rating System
French market - 2006
-2
-1
0
1
2
-2
-1
0
1
2
-1.5
-1
-0.5
0
0.5
1
cr
eb
Score (EBIT, Current ratio)
MOGA
Multiobjective Genetic Algorithms
MOGA ā€“ feature selection
S-ISOMAP ā€“ manifold learning
The idea behind it
Other approaches
ā€¢ SVM+ - domain knowledge SVMs
ā€¢ RVM ā€“ probabilistic SVMs
ā€¢ NMF ā€“ Non-negative Matrix
Factorization
ā€¢ Genetic Programming
ā€¢ ā€¦
The Power of Social Network
Analysis
Bad Rank Algorithm
Where are the bad guys?
Bad Rank for Fraud Detection
Results with Semi-supervised Learning
Networks Analysis
A world of possibilities
ā€¢ Identify critical nodes / links / clusters
ā€¢ Detailed information of dynamics
ā€¢ Stability / robustness of system
ā€¢ Information / crisis Propagation
ā€¢ Stress tests
Team
JoĆ£o Carvalho das Neves
Professor of
Management, ISEG.
Ph.D. in Business
Administration,
Manchester Business
School
Armando Vieira
Professor of Physics, &
entrepreneur. Ph.D. in
Physics and researcher
in Artificial Intelligence
Bernardete Ribeiro
Associate Professor
of Computer
Science, University
Coimbra,
researcher at
CISUC.
Tiago Marques
Marketing and
Business
Consultant,
E-Business
Specialist,
Director of
Research
Business
Director
IT Researcher Marketing
10+ years experience in AI
25 years experience in Credit Risk & Financial Analysis
15 years of marketing experience
What do banks need in credit
management?
Efficiency Accuracy
Savings of Capital ā€“ Basel requirements
This is a highly regulated industry with detailed and focused regulators
What do they get?
Boosting the accuracy of credit risk methodologies will lead to considerable gains for banks
Source: Issue 2 ofĀ NPLEurope, a publication overing non-performing loan
(NPL) markets in Europe and the United Kingdom (UK).,
PriceWaterhouseCoopers
Non-performing loans - Europe
0
50
100
150
200
250
Germany UK Spain Italy Russia Greece
2008
2009
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
2005 2006 2007 2008 2009
% Corporate Debt Default -
Portugal
BillionsofEUR
NPL(%)
Source: Bank of Portugal
AIRES Solution
AIRES.dei.uc.pt

More Related Content

Viewers also liked

8 9 forecasting of financial statements
8 9   forecasting of financial statements8 9   forecasting of financial statements
8 9 forecasting of financial statementsJohn McSherry
Ā 
Manifold learning for credit risk assessment
Manifold learning for credit risk assessment Manifold learning for credit risk assessment
Manifold learning for credit risk assessment Armando Vieira
Ā 
Artificial neural networks for ion beam analysis
Artificial neural networks for ion beam analysisArtificial neural networks for ion beam analysis
Artificial neural networks for ion beam analysisArmando Vieira
Ā 
Credit risk meetup
Credit risk meetupCredit risk meetup
Credit risk meetupQuantUniversity
Ā 
Non Performing Loans (NPLā€˜s) ā€“ how to handle and optimize
Non Performing Loans (NPLā€˜s) ā€“ how to handle and optimizeNon Performing Loans (NPLā€˜s) ā€“ how to handle and optimize
Non Performing Loans (NPLā€˜s) ā€“ how to handle and optimizeLĆ”szlĆ³ Ɓrvai
Ā 
Instilling the Right Credit Risk Culture
Instilling the Right Credit Risk CultureInstilling the Right Credit Risk Culture
Instilling the Right Credit Risk CultureLibby Bierman
Ā 
ppt spatial data
ppt spatial datappt spatial data
ppt spatial dataRahul Kumar
Ā 
Supply Chain Risk
Supply Chain RiskSupply Chain Risk
Supply Chain RiskJan Husdal
Ā 
Supply Chain Risk Management
Supply Chain Risk ManagementSupply Chain Risk Management
Supply Chain Risk ManagementAnand Subramaniam
Ā 

Viewers also liked (9)

8 9 forecasting of financial statements
8 9   forecasting of financial statements8 9   forecasting of financial statements
8 9 forecasting of financial statements
Ā 
Manifold learning for credit risk assessment
Manifold learning for credit risk assessment Manifold learning for credit risk assessment
Manifold learning for credit risk assessment
Ā 
Artificial neural networks for ion beam analysis
Artificial neural networks for ion beam analysisArtificial neural networks for ion beam analysis
Artificial neural networks for ion beam analysis
Ā 
Credit risk meetup
Credit risk meetupCredit risk meetup
Credit risk meetup
Ā 
Non Performing Loans (NPLā€˜s) ā€“ how to handle and optimize
Non Performing Loans (NPLā€˜s) ā€“ how to handle and optimizeNon Performing Loans (NPLā€˜s) ā€“ how to handle and optimize
Non Performing Loans (NPLā€˜s) ā€“ how to handle and optimize
Ā 
Instilling the Right Credit Risk Culture
Instilling the Right Credit Risk CultureInstilling the Right Credit Risk Culture
Instilling the Right Credit Risk Culture
Ā 
ppt spatial data
ppt spatial datappt spatial data
ppt spatial data
Ā 
Supply Chain Risk
Supply Chain RiskSupply Chain Risk
Supply Chain Risk
Ā 
Supply Chain Risk Management
Supply Chain Risk ManagementSupply Chain Risk Management
Supply Chain Risk Management
Ā 

Similar to Credit risk with neural networks bankruptcy prediction machine learning

Fractal Labs Capability Set
Fractal Labs Capability SetFractal Labs Capability Set
Fractal Labs Capability SetMark Young
Ā 
Personal Loan Risk Assessment
Personal Loan Risk Assessment Personal Loan Risk Assessment
Personal Loan Risk Assessment Kunal Kashyap
Ā 
Salesforce SMIF FINAL Presntation
Salesforce SMIF FINAL PresntationSalesforce SMIF FINAL Presntation
Salesforce SMIF FINAL PresntationGabriel E. Garcia
Ā 
Temenos Insight Risk
Temenos Insight RiskTemenos Insight Risk
Temenos Insight Riskahmedzafar
Ā 
Operation var (ama) con0529e
Operation var (ama) con0529eOperation var (ama) con0529e
Operation var (ama) con0529eChipo Nyachiwowa
Ā 
The future of the OTC Derivative Market - Eugene stanfield
The future of the OTC Derivative Market - Eugene stanfieldThe future of the OTC Derivative Market - Eugene stanfield
The future of the OTC Derivative Market - Eugene stanfieldLĆ”szlĆ³ Ɓrvai
Ā 
Business Case Calculator for DevOps Initiatives - Leading credit card service...
Business Case Calculator for DevOps Initiatives - Leading credit card service...Business Case Calculator for DevOps Initiatives - Leading credit card service...
Business Case Calculator for DevOps Initiatives - Leading credit card service...Capgemini
Ā 
Gaining a Competitive Advantage using Analytics to Optimize your Digital Mark...
Gaining a Competitive Advantage using Analytics to Optimize your Digital Mark...Gaining a Competitive Advantage using Analytics to Optimize your Digital Mark...
Gaining a Competitive Advantage using Analytics to Optimize your Digital Mark...Abbie Wong
Ā 
Wooing the Best Bank Deposit Customers
Wooing the Best Bank Deposit CustomersWooing the Best Bank Deposit Customers
Wooing the Best Bank Deposit CustomersLucinda Linde
Ā 
Leveraging KPIā€™s to Maximize the ROI of Support
Leveraging KPIā€™s to Maximize the ROI of Support Leveraging KPIā€™s to Maximize the ROI of Support
Leveraging KPIā€™s to Maximize the ROI of Support MetricNet
Ā 
Edge_Strategy_Opentext_Supplier_Risk_Management.pdf
Edge_Strategy_Opentext_Supplier_Risk_Management.pdfEdge_Strategy_Opentext_Supplier_Risk_Management.pdf
Edge_Strategy_Opentext_Supplier_Risk_Management.pdfSubrat Kumar Dash
Ā 
Unleashing the Enormous Power of Service Desk KPIs
Unleashing the Enormous Power of Service Desk KPIsUnleashing the Enormous Power of Service Desk KPIs
Unleashing the Enormous Power of Service Desk KPIsMetricNet
Ā 
Artificial Intelligence high ROI case studies from around the world: approach...
Artificial Intelligence high ROI case studies from around the world: approach...Artificial Intelligence high ROI case studies from around the world: approach...
Artificial Intelligence high ROI case studies from around the world: approach...Data Driven Innovation
Ā 
Microsoft analysis.pptx
Microsoft analysis.pptxMicrosoft analysis.pptx
Microsoft analysis.pptxMartin373349
Ā 
Credit Risk Management Presentation
Credit Risk Management PresentationCredit Risk Management Presentation
Credit Risk Management PresentationSumant Palwankar
Ā 
Level 3
Level 3Level 3
Level 3GWC GROUP
Ā 

Similar to Credit risk with neural networks bankruptcy prediction machine learning (20)

Leuven
LeuvenLeuven
Leuven
Ā 
Fractal Labs Capability Set
Fractal Labs Capability SetFractal Labs Capability Set
Fractal Labs Capability Set
Ā 
Personal Loan Risk Assessment
Personal Loan Risk Assessment Personal Loan Risk Assessment
Personal Loan Risk Assessment
Ā 
Salesforce SMIF FINAL Presntation
Salesforce SMIF FINAL PresntationSalesforce SMIF FINAL Presntation
Salesforce SMIF FINAL Presntation
Ā 
Temenos Insight Risk
Temenos Insight RiskTemenos Insight Risk
Temenos Insight Risk
Ā 
Operation var (ama) con0529e
Operation var (ama) con0529eOperation var (ama) con0529e
Operation var (ama) con0529e
Ā 
The future of the OTC Derivative Market - Eugene stanfield
The future of the OTC Derivative Market - Eugene stanfieldThe future of the OTC Derivative Market - Eugene stanfield
The future of the OTC Derivative Market - Eugene stanfield
Ā 
Business Case Calculator for DevOps Initiatives - Leading credit card service...
Business Case Calculator for DevOps Initiatives - Leading credit card service...Business Case Calculator for DevOps Initiatives - Leading credit card service...
Business Case Calculator for DevOps Initiatives - Leading credit card service...
Ā 
Gaining a Competitive Advantage using Analytics to Optimize your Digital Mark...
Gaining a Competitive Advantage using Analytics to Optimize your Digital Mark...Gaining a Competitive Advantage using Analytics to Optimize your Digital Mark...
Gaining a Competitive Advantage using Analytics to Optimize your Digital Mark...
Ā 
Wooing the Best Bank Deposit Customers
Wooing the Best Bank Deposit CustomersWooing the Best Bank Deposit Customers
Wooing the Best Bank Deposit Customers
Ā 
Leveraging KPIā€™s to Maximize the ROI of Support
Leveraging KPIā€™s to Maximize the ROI of Support Leveraging KPIā€™s to Maximize the ROI of Support
Leveraging KPIā€™s to Maximize the ROI of Support
Ā 
TALEO_Reporting_Global_VF
TALEO_Reporting_Global_VFTALEO_Reporting_Global_VF
TALEO_Reporting_Global_VF
Ā 
Edge_Strategy_Opentext_Supplier_Risk_Management.pdf
Edge_Strategy_Opentext_Supplier_Risk_Management.pdfEdge_Strategy_Opentext_Supplier_Risk_Management.pdf
Edge_Strategy_Opentext_Supplier_Risk_Management.pdf
Ā 
RiskMngForum_MyPresentation_Istanbul_Summary
RiskMngForum_MyPresentation_Istanbul_SummaryRiskMngForum_MyPresentation_Istanbul_Summary
RiskMngForum_MyPresentation_Istanbul_Summary
Ā 
Unleashing the Enormous Power of Service Desk KPIs
Unleashing the Enormous Power of Service Desk KPIsUnleashing the Enormous Power of Service Desk KPIs
Unleashing the Enormous Power of Service Desk KPIs
Ā 
Six Sigma.pptx
Six Sigma.pptxSix Sigma.pptx
Six Sigma.pptx
Ā 
Artificial Intelligence high ROI case studies from around the world: approach...
Artificial Intelligence high ROI case studies from around the world: approach...Artificial Intelligence high ROI case studies from around the world: approach...
Artificial Intelligence high ROI case studies from around the world: approach...
Ā 
Microsoft analysis.pptx
Microsoft analysis.pptxMicrosoft analysis.pptx
Microsoft analysis.pptx
Ā 
Credit Risk Management Presentation
Credit Risk Management PresentationCredit Risk Management Presentation
Credit Risk Management Presentation
Ā 
Level 3
Level 3Level 3
Level 3
Ā 

More from Armando Vieira

Improving Insurance Risk Prediction with Generative Adversarial Networks (GANs)
Improving Insurance  Risk Prediction with Generative Adversarial Networks (GANs)Improving Insurance  Risk Prediction with Generative Adversarial Networks (GANs)
Improving Insurance Risk Prediction with Generative Adversarial Networks (GANs)Armando Vieira
Ā 
Predicting online user behaviour using deep learning algorithms
Predicting online user behaviour using deep learning algorithmsPredicting online user behaviour using deep learning algorithms
Predicting online user behaviour using deep learning algorithmsArmando Vieira
Ā 
Boosting conversion rates on ecommerce using deep learning algorithms
Boosting conversion rates on ecommerce using deep learning algorithmsBoosting conversion rates on ecommerce using deep learning algorithms
Boosting conversion rates on ecommerce using deep learning algorithmsArmando Vieira
Ā 
Seasonality effects on second hand cars sales
Seasonality effects on second hand cars salesSeasonality effects on second hand cars sales
Seasonality effects on second hand cars salesArmando Vieira
Ā 
Visualizations of high dimensional data using R and Shiny
Visualizations of high dimensional data using R and ShinyVisualizations of high dimensional data using R and Shiny
Visualizations of high dimensional data using R and ShinyArmando Vieira
Ā 
Dl2 computing gpu
Dl2 computing gpuDl2 computing gpu
Dl2 computing gpuArmando Vieira
Ā 
Dl1 deep learning_algorithms
Dl1 deep learning_algorithmsDl1 deep learning_algorithms
Dl1 deep learning_algorithmsArmando Vieira
Ā 
Extracting Knowledge from Pydata London 2015
Extracting Knowledge from Pydata London 2015Extracting Knowledge from Pydata London 2015
Extracting Knowledge from Pydata London 2015Armando Vieira
Ā 
Hidden Layer Leraning Vector Quantizatio
Hidden Layer Leraning Vector Quantizatio Hidden Layer Leraning Vector Quantizatio
Hidden Layer Leraning Vector Quantizatio Armando Vieira
Ā 
machine learning in the age of big data: new approaches and business applicat...
machine learning in the age of big data: new approaches and business applicat...machine learning in the age of big data: new approaches and business applicat...
machine learning in the age of big data: new approaches and business applicat...Armando Vieira
Ā 
Neural Networks and Genetic Algorithms Multiobjective acceleration
Neural Networks and Genetic Algorithms Multiobjective accelerationNeural Networks and Genetic Algorithms Multiobjective acceleration
Neural Networks and Genetic Algorithms Multiobjective accelerationArmando Vieira
Ā 
Optimization of digital marketing campaigns
Optimization of digital marketing campaignsOptimization of digital marketing campaigns
Optimization of digital marketing campaignsArmando Vieira
Ā 
Online democracy Armando Vieira
Online democracy Armando VieiraOnline democracy Armando Vieira
Online democracy Armando VieiraArmando Vieira
Ā 
Invtur conference aveiro 2010
Invtur conference aveiro 2010Invtur conference aveiro 2010
Invtur conference aveiro 2010Armando Vieira
Ā 
Tourism with recomendation systems
Tourism with recomendation systemsTourism with recomendation systems
Tourism with recomendation systemsArmando Vieira
Ā 
Requiem pelo ensino
Requiem pelo ensino Requiem pelo ensino
Requiem pelo ensino Armando Vieira
Ā 
Pattern recognition
Pattern recognitionPattern recognition
Pattern recognitionArmando Vieira
Ā 
Key ratios for financial analysis
Key ratios for financial analysisKey ratios for financial analysis
Key ratios for financial analysisArmando Vieira
Ā 

More from Armando Vieira (20)

Improving Insurance Risk Prediction with Generative Adversarial Networks (GANs)
Improving Insurance  Risk Prediction with Generative Adversarial Networks (GANs)Improving Insurance  Risk Prediction with Generative Adversarial Networks (GANs)
Improving Insurance Risk Prediction with Generative Adversarial Networks (GANs)
Ā 
Predicting online user behaviour using deep learning algorithms
Predicting online user behaviour using deep learning algorithmsPredicting online user behaviour using deep learning algorithms
Predicting online user behaviour using deep learning algorithms
Ā 
Boosting conversion rates on ecommerce using deep learning algorithms
Boosting conversion rates on ecommerce using deep learning algorithmsBoosting conversion rates on ecommerce using deep learning algorithms
Boosting conversion rates on ecommerce using deep learning algorithms
Ā 
Seasonality effects on second hand cars sales
Seasonality effects on second hand cars salesSeasonality effects on second hand cars sales
Seasonality effects on second hand cars sales
Ā 
Visualizations of high dimensional data using R and Shiny
Visualizations of high dimensional data using R and ShinyVisualizations of high dimensional data using R and Shiny
Visualizations of high dimensional data using R and Shiny
Ā 
Dl2 computing gpu
Dl2 computing gpuDl2 computing gpu
Dl2 computing gpu
Ā 
Dl1 deep learning_algorithms
Dl1 deep learning_algorithmsDl1 deep learning_algorithms
Dl1 deep learning_algorithms
Ā 
Extracting Knowledge from Pydata London 2015
Extracting Knowledge from Pydata London 2015Extracting Knowledge from Pydata London 2015
Extracting Knowledge from Pydata London 2015
Ā 
Hidden Layer Leraning Vector Quantizatio
Hidden Layer Leraning Vector Quantizatio Hidden Layer Leraning Vector Quantizatio
Hidden Layer Leraning Vector Quantizatio
Ā 
machine learning in the age of big data: new approaches and business applicat...
machine learning in the age of big data: new approaches and business applicat...machine learning in the age of big data: new approaches and business applicat...
machine learning in the age of big data: new approaches and business applicat...
Ā 
Neural Networks and Genetic Algorithms Multiobjective acceleration
Neural Networks and Genetic Algorithms Multiobjective accelerationNeural Networks and Genetic Algorithms Multiobjective acceleration
Neural Networks and Genetic Algorithms Multiobjective acceleration
Ā 
Optimization of digital marketing campaigns
Optimization of digital marketing campaignsOptimization of digital marketing campaigns
Optimization of digital marketing campaigns
Ā 
Online democracy Armando Vieira
Online democracy Armando VieiraOnline democracy Armando Vieira
Online democracy Armando Vieira
Ā 
Invtur conference aveiro 2010
Invtur conference aveiro 2010Invtur conference aveiro 2010
Invtur conference aveiro 2010
Ā 
Tourism with recomendation systems
Tourism with recomendation systemsTourism with recomendation systems
Tourism with recomendation systems
Ā 
Credit iconip
Credit iconipCredit iconip
Credit iconip
Ā 
Requiem pelo ensino
Requiem pelo ensino Requiem pelo ensino
Requiem pelo ensino
Ā 
Eurogen v
Eurogen vEurogen v
Eurogen v
Ā 
Pattern recognition
Pattern recognitionPattern recognition
Pattern recognition
Ā 
Key ratios for financial analysis
Key ratios for financial analysisKey ratios for financial analysis
Key ratios for financial analysis
Ā 

Recently uploaded

EMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWER
EMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWEREMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWER
EMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWERMadyBayot
Ā 
Exploring Multimodal Embeddings with Milvus
Exploring Multimodal Embeddings with MilvusExploring Multimodal Embeddings with Milvus
Exploring Multimodal Embeddings with MilvusZilliz
Ā 
CNIC Information System with Pakdata Cf In Pakistan
CNIC Information System with Pakdata Cf In PakistanCNIC Information System with Pakdata Cf In Pakistan
CNIC Information System with Pakdata Cf In Pakistandanishmna97
Ā 
Architecting Cloud Native Applications
Architecting Cloud Native ApplicationsArchitecting Cloud Native Applications
Architecting Cloud Native ApplicationsWSO2
Ā 
AXA XL - Insurer Innovation Award Americas 2024
AXA XL - Insurer Innovation Award Americas 2024AXA XL - Insurer Innovation Award Americas 2024
AXA XL - Insurer Innovation Award Americas 2024The Digital Insurer
Ā 
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers:  A Deep Dive into Serverless Spatial Data and FMECloud Frontiers:  A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FMESafe Software
Ā 
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, AdobeApidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobeapidays
Ā 
Corporate and higher education May webinar.pptx
Corporate and higher education May webinar.pptxCorporate and higher education May webinar.pptx
Corporate and higher education May webinar.pptxRustici Software
Ā 
Apidays New York 2024 - Passkeys: Developing APIs to enable passwordless auth...
Apidays New York 2024 - Passkeys: Developing APIs to enable passwordless auth...Apidays New York 2024 - Passkeys: Developing APIs to enable passwordless auth...
Apidays New York 2024 - Passkeys: Developing APIs to enable passwordless auth...apidays
Ā 
Manulife - Insurer Transformation Award 2024
Manulife - Insurer Transformation Award 2024Manulife - Insurer Transformation Award 2024
Manulife - Insurer Transformation Award 2024The Digital Insurer
Ā 
"I see eyes in my soup": How Delivery Hero implemented the safety system for ...
"I see eyes in my soup": How Delivery Hero implemented the safety system for ..."I see eyes in my soup": How Delivery Hero implemented the safety system for ...
"I see eyes in my soup": How Delivery Hero implemented the safety system for ...Zilliz
Ā 
TrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data DiscoveryTrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data DiscoveryTrustArc
Ā 
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers:  A Deep Dive into Serverless Spatial Data and FMECloud Frontiers:  A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FMESafe Software
Ā 
DBX First Quarter 2024 Investor Presentation
DBX First Quarter 2024 Investor PresentationDBX First Quarter 2024 Investor Presentation
DBX First Quarter 2024 Investor PresentationDropbox
Ā 
Strategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
Strategize a Smooth Tenant-to-tenant Migration and Copilot TakeoffStrategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
Strategize a Smooth Tenant-to-tenant Migration and Copilot Takeoffsammart93
Ā 
Strategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a FresherStrategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a FresherRemote DBA Services
Ā 
Finding Java's Hidden Performance Traps @ DevoxxUK 2024
Finding Java's Hidden Performance Traps @ DevoxxUK 2024Finding Java's Hidden Performance Traps @ DevoxxUK 2024
Finding Java's Hidden Performance Traps @ DevoxxUK 2024Victor Rentea
Ā 
Apidays New York 2024 - The Good, the Bad and the Governed by David O'Neill, ...
Apidays New York 2024 - The Good, the Bad and the Governed by David O'Neill, ...Apidays New York 2024 - The Good, the Bad and the Governed by David O'Neill, ...
Apidays New York 2024 - The Good, the Bad and the Governed by David O'Neill, ...apidays
Ā 
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemkeProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemkeProduct Anonymous
Ā 

Recently uploaded (20)

+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
Ā 
EMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWER
EMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWEREMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWER
EMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWER
Ā 
Exploring Multimodal Embeddings with Milvus
Exploring Multimodal Embeddings with MilvusExploring Multimodal Embeddings with Milvus
Exploring Multimodal Embeddings with Milvus
Ā 
CNIC Information System with Pakdata Cf In Pakistan
CNIC Information System with Pakdata Cf In PakistanCNIC Information System with Pakdata Cf In Pakistan
CNIC Information System with Pakdata Cf In Pakistan
Ā 
Architecting Cloud Native Applications
Architecting Cloud Native ApplicationsArchitecting Cloud Native Applications
Architecting Cloud Native Applications
Ā 
AXA XL - Insurer Innovation Award Americas 2024
AXA XL - Insurer Innovation Award Americas 2024AXA XL - Insurer Innovation Award Americas 2024
AXA XL - Insurer Innovation Award Americas 2024
Ā 
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers:  A Deep Dive into Serverless Spatial Data and FMECloud Frontiers:  A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FME
Ā 
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, AdobeApidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Ā 
Corporate and higher education May webinar.pptx
Corporate and higher education May webinar.pptxCorporate and higher education May webinar.pptx
Corporate and higher education May webinar.pptx
Ā 
Apidays New York 2024 - Passkeys: Developing APIs to enable passwordless auth...
Apidays New York 2024 - Passkeys: Developing APIs to enable passwordless auth...Apidays New York 2024 - Passkeys: Developing APIs to enable passwordless auth...
Apidays New York 2024 - Passkeys: Developing APIs to enable passwordless auth...
Ā 
Manulife - Insurer Transformation Award 2024
Manulife - Insurer Transformation Award 2024Manulife - Insurer Transformation Award 2024
Manulife - Insurer Transformation Award 2024
Ā 
"I see eyes in my soup": How Delivery Hero implemented the safety system for ...
"I see eyes in my soup": How Delivery Hero implemented the safety system for ..."I see eyes in my soup": How Delivery Hero implemented the safety system for ...
"I see eyes in my soup": How Delivery Hero implemented the safety system for ...
Ā 
TrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data DiscoveryTrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
Ā 
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers:  A Deep Dive into Serverless Spatial Data and FMECloud Frontiers:  A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FME
Ā 
DBX First Quarter 2024 Investor Presentation
DBX First Quarter 2024 Investor PresentationDBX First Quarter 2024 Investor Presentation
DBX First Quarter 2024 Investor Presentation
Ā 
Strategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
Strategize a Smooth Tenant-to-tenant Migration and Copilot TakeoffStrategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
Strategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
Ā 
Strategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a FresherStrategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a Fresher
Ā 
Finding Java's Hidden Performance Traps @ DevoxxUK 2024
Finding Java's Hidden Performance Traps @ DevoxxUK 2024Finding Java's Hidden Performance Traps @ DevoxxUK 2024
Finding Java's Hidden Performance Traps @ DevoxxUK 2024
Ā 
Apidays New York 2024 - The Good, the Bad and the Governed by David O'Neill, ...
Apidays New York 2024 - The Good, the Bad and the Governed by David O'Neill, ...Apidays New York 2024 - The Good, the Bad and the Governed by David O'Neill, ...
Apidays New York 2024 - The Good, the Bad and the Governed by David O'Neill, ...
Ā 
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemkeProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
Ā 

Credit risk with neural networks bankruptcy prediction machine learning

  • 1. Credit Risk with AI tools The old, the new and the unexpected ARMANDO VIEIRA Armandosvieira.wordpress.com
  • 2. Customer fails to pay Losing money Wrong Strategy Change in market prices Processing failures and frauds Regulatory compliance Customer fails to pay Losing money Wrong Strategy Change in market prices Processing failures and frauds Regulatory compliance RISK
  • 4. A statistical means of providing a quantifiable risk factor for a given applicant. Credit scoring is a process whereby information provided is converted into numbers to arrive at a score. The objective is to forecast future performance from past behavior of clients (SME or individuals). Credit scoring are used in many areas of industries: Banking Decision Models Finance Insurance Retail Telecommunications What is Credit Scoring?
  • 5.
  • 6. ā€¢ Predict financial distress of private companies one year ahead based on account balance sheet from previous years. ā€¢ Enventualy the probability to become so. ā€¢ Obtain reliable data from up to 5 previous years before failure ā€¢ Classify and release warning signs Bankruptcy prediction problem
  • 7. The curse of dimensionality Problems ā€¢ Sparness of the search space ā€¢ Presence of Irrelevant Features ā€¢ Poor generalization of Learning Machine ā€¢ Exceptions difficult to identify Solutions ā€¢ Dimensionality reduction: feature selection ā€¢ Constrain the complexity of the Learning Machine
  • 8. The Diane Database ā€¢ Financial statements of French companies, initially of 60,000 industrial French companies, for the years of 2002 to 2006, with at least 10 employees ā€¢ 3,000 were declared bankrupted in 2007 or presented a ā€¢ restructuring plan 30 financial ratios which allow the description of firms in terms of the financial strength, liquidity, solvability, productivity of labor and capital, margins, net profitability and return on investment
  • 9. The inputs Number of employees Net Current Assets/Turnover (days) Financial Debt / Capital Employed (%) Working Capital Needs / Turnover (%) Capital Employed / Fixed Assets Export (%) Depreciation of Tangible Assets (%) Value added per employee Working capital / current assets Total Assets / Turnover Current ratio Operating Profit Margin (%) Liquidity ratio Net Profit Margin (%) Stock Turnover days Added Value Margin (%) Collection period Part of Employees (%) Credit Period Return on Capital Employed (%) Turnover per Employee Return on Total Assets (%) Interest / Turnover EBIT Margin (%) Debt Period (days) EBITDA Margin (%) Financial Debt / Equity (%) Cashflow / Turnover (%) Financial Debt / Cashflow Working Capital / Turnover (days)
  • 10. Hard problem 0 2 4 6 3 4 5 6 7 Class 0 Class 1 Ī» 1 Ī» 2 First two principal component from PCA
  • 11. How HLVQ-C works 0 0.5 1.0 1.5 0 0.5 1.0 1.5 Class 0 Class 1 After Before ? d2 d1 X Y
  • 12. DIANE 1 (error%) Model Error I Error II Total MDA SVM MLP HLVQ-C 26.4 17.6 25.7 11.1 21.0 12.2 13.1 10.6 23.7 14.8 19.4 10.8
  • 13. DIANE 1 - HLVQC Results Method Classification Weighted Efficiency (%) Z-score (Altman) 62.7 Best Discriminant 66.1 MLP 71.4 OurMethod 84.1 Source: Vieira, A.S., Neves, J.C.: Improving Bankruptcy Prediction with Hidden Layer Learning. Vector Quantization. European Accounting Review, 15 (2), 253-271 (2006).
  • 15. Results I ā€“ 30 days into arrears Classifier Accuracy (%) Type I Type II G Logistic 66.3 27.3 40.1 54.8 MLP 67.5 8.1 57.1 61.1 SVM 64.9 35.6 34.6 52.3 AdaboostM1 69.0 12.6 49.4 55.7 HLVQ-C 72.6 5.3 49.5 52.3
  • 16. Results I ā€“ 60 days into arrears Classifier Accuracy Type I Type II G Logistic 81.2 48.2 11.0 21.2 MLP 82.3 57.4 9.1 20.1 SVM 83.3 38.1 12.4 19.3 AdaboostM1 84.1 45.7 8.0 14.7 HLVQ-C 86.5 48.3 6.2 11.9
  • 17. DIANE II (2002 ā€“ 2007) ā€¢ More data ā€¢ Longer history ā€¢ More features
  • 18. Year 2006 Classifier Accuracy Type I Type II Logistic 91.25 6.33 11.17 MLP 91.17 6.33 11.33 C-SVM 92.42 5.16 10.00 AdaboostM1 89.75 8.16 12.33 Year 2005 Classifier Accuracy Type I Type II Logistic 79.92 19.50 20.67 MLP 75.83 24.50 23.83 C-SVM 80.00 21.17 18.83 AdaboostM1 78.17 20.50 23.17 Results
  • 19. How useful? [ ]mexexNV III )1()1( āˆ’āˆ’āˆ’=Ī· ļ£·ļ£· ļ£ø ļ£¶ ļ£¬ļ£¬ ļ£­ ļ£« āˆ’ >> āˆ’ I II e e mmG x x 11
  • 22.
  • 23.
  • 24.
  • 27. MOGA ā€“ feature selection
  • 28.
  • 29.
  • 32.
  • 33.
  • 34.
  • 35. Other approaches ā€¢ SVM+ - domain knowledge SVMs ā€¢ RVM ā€“ probabilistic SVMs ā€¢ NMF ā€“ Non-negative Matrix Factorization ā€¢ Genetic Programming ā€¢ ā€¦
  • 36. The Power of Social Network Analysis
  • 38. Where are the bad guys?
  • 39. Bad Rank for Fraud Detection
  • 41. Networks Analysis A world of possibilities ā€¢ Identify critical nodes / links / clusters ā€¢ Detailed information of dynamics ā€¢ Stability / robustness of system ā€¢ Information / crisis Propagation ā€¢ Stress tests
  • 42.
  • 43.
  • 44. Team JoĆ£o Carvalho das Neves Professor of Management, ISEG. Ph.D. in Business Administration, Manchester Business School Armando Vieira Professor of Physics, & entrepreneur. Ph.D. in Physics and researcher in Artificial Intelligence Bernardete Ribeiro Associate Professor of Computer Science, University Coimbra, researcher at CISUC. Tiago Marques Marketing and Business Consultant, E-Business Specialist, Director of Research Business Director IT Researcher Marketing 10+ years experience in AI 25 years experience in Credit Risk & Financial Analysis 15 years of marketing experience
  • 45. What do banks need in credit management? Efficiency Accuracy Savings of Capital ā€“ Basel requirements This is a highly regulated industry with detailed and focused regulators
  • 46. What do they get? Boosting the accuracy of credit risk methodologies will lead to considerable gains for banks Source: Issue 2 ofĀ NPLEurope, a publication overing non-performing loan (NPL) markets in Europe and the United Kingdom (UK)., PriceWaterhouseCoopers Non-performing loans - Europe 0 50 100 150 200 250 Germany UK Spain Italy Russia Greece 2008 2009 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 2005 2006 2007 2008 2009 % Corporate Debt Default - Portugal BillionsofEUR NPL(%) Source: Bank of Portugal

Editor's Notes

  1. the banking industry is a highly regulated industry with detailed and focused regulators Fast, fully adaptable, performance and accuracy Commercial Benefits Cost Reduction Investor Scale NegĆ³cio que irĆ” permanecer com alta procura ROI Of the team An experienced team, where the whole is far greater than the sum of its parts
  2. Boosting the accuracy of credit risk methodologies used by banks and financial institutions may lead to considerable gains. Default rate in Portugal has more than double in the past 5 years Similary in Europe NPL increase by over 25%, many as much as 50% 620 billion euros in 2009 For example, improving the accuracy of credit risk assessment models by only 1% may lead to a gain in banking sector of about 50 million Euros - in Portugal alone