Bridging the Academic–Practitioner Divide in Credit Risk Modeling Vadim Melnitchouk, Metropoliten State University, Saint Paul, MN, US.

Slides:



Advertisements
Similar presentations
Presentation to the NABE Corporate Planning Roundtable Teleconference Ardavan Mobasheri Head of AIG Global Economics.
Advertisements

Value-at-Risk: A Risk Estimating Tool for Management
Comparing Time Series, Neural Nets and Probability Models for New Product Trial Forecasting Eugene Brusilovskiy Ka Lok Lee These slides are based on the.
AACEI Contingency Forum Contingency Management
Understanding Private Loans Default Prevention. Agenda  Essential loan language  Variable rate language ♦ Types of indexes  Language for all types.
Credit Risk Assessment of Corporate Sector in Croatia Saša Cerovac, Lana Ivičić Croatian National Bank Financial Stability Department.
An Analysis on estimating Funds Requirements Presented By : Saurabh Kumar Sinha 2009PGP049 Saurabh Patawari 2009PGP050 Siddharth Shankar Prasad 2009PGP051.
 New reforms have been passed to modernize the Swiss electricity sector.  Legal (transmission and generation are separate legal entities) and functional.
©2013 Experian Ltd. All rights reserved. No part of this copyrighted work may be reproduced, modified, or distributed in any form or manner without the.
Slide: 1 Stress tests A theoretical exercise or do they really work? TopQuants Autumn Event 2013 By Robert Daniels Partner at Capstone Financial Industry.
Two Applied Papers on Measurement Error in Wages Downward nominal wage flexibility– real or measurement error? Impact of Non-Classical Measurement Error.
An Overview of Machine Learning
Part 24: Bayesian Estimation 24-1/35 Econometrics I Professor William Greene Stern School of Business Department of Economics.
A Training Session by National Community Capital Association 1 Risk Management for Loan Programs RESNA Alternative Financing and Telework Loan Programs.
Cox Model With Intermitten and Error-Prone Covariate Observation Yury Gubman PhD thesis in Statistics Supervisors: Prof. David Zucker, Prof. Orly Manor.
STCPM title A model of bank price and nonprice competition with endogenous expected loan losses Filipa Lima Paulo Soares de Pinho Emerging Scholars in.
Part 21: Hazard Models [1/29] Econometric Analysis of Panel Data William Greene Department of Economics Stern School of Business.
Simultaneous Forecasting of Non-stationary conditional Mean & Variance Speaker: Andrey Torzhkov September 25 th, 2006.
End of Chapter 8 Neil Weisenfeld March 28, 2005.
Jiahan Li Assistant professor of Statistics University of Notre Dame
Demand Planning: Forecasting and Demand Management
A Global Macroeconomic Forecasting Model for the Philippines Ruperto Majuca, Ph.D (Illinois), J.D. De La Salle University, Manila 51 st Philippine Economic.
MILAN| 23 June 2011 Success factors of models used for supporting sales ‐ related allocation decisions - Sönke Albers Professor of Marketing and Innovation.
These views are my own and do not necessarily represent the views of the Federal Reserve Bank of New York or the Federal Reserve System Underestimating.
Modeling Impacts from Current Expected Credit Loss Framework
Chapter 2 – Business Forecasting Takesh Luckho. What is Business Forecasting?  Forecasting is about predicting the future as accurately as possible,
Modelling Credit Risk Croatian Quants Day Vančo Balen
LECTURE 22 VAR 1. Methods of calculating VAR (Cont.) Correlation method is conceptually simple and easy to apply; it only requires the mean returns and.
Efficiency Measurement William Greene Stern School of Business New York University.
© 2012 Cengage Learning. Residential Mortgage Lending: Principles and Practices, 6e Chapter 3 Role of Residential Mortgage Lending in the Economy.
Discussion of “Foreclosures In Ohio: Does Lender Type Matter?” Robert B. Avery January 2, 2009.
1 QUANTITATIVE RISK MANAGEMENT AT ABN AMRO Jan Sijbrand January 14th, 2000.
Chapter 24 Strategies and Rules for Monetary Policy Introduction to Economics (Combined Version) 5th Edition.
Discussion Resolution Policy and the Cost of Bank Failures.
1 Mortgage Defaults and Foreclosures: Recent Trends and Associated Economic and Market Developments Randy Fasnacht U.S. Government Accountability Office.
R. Bhar, A. G. Malliariswww.bhar.id.au1 Deviation of US Equity Return Are They Determined by Fundamental or Behavioral Variables? R. Bhar (UNSW), A. G.
The Demand for Home Equity Loans at Bank X* An MBA 555 Project Laura Brown Richard Brown Jason Vanderploeg *bank name withheld for proprietary reasons.
ONE COUNTRY TWO ECONOMIES Bill Dunkelberg, Chief Economist – NFIB William J. Dennis, Jr., Pinch Hitter – NFIB Richmond Fed Credit Markets Symposium April,
MIS An Economic Analysis of Software Market with Risk-Sharing Contract Byung Cho Kim Pei-Yu Chen Tridas Mukhopadhyay Tepper School of Business Carnegie.
Estimating the Predictive Distribution for Loss Reserve Models Glenn Meyers Casualty Loss Reserve Seminar September 12, 2006.
The Anatomy of Household Debt Build Up: What Are the Implications for the Financial Stability in Croatia? Ivana Herceg and Vedran Šošić* *Views expressed.
Estimating Credit Exposure and Economic Capital Using Monte Carlo Simulation Ronald Lagnado Vice President, MKIRisk IPAM Conference on Financial Mathematics.
Discrete Choice Modeling William Greene Stern School of Business New York University.
Reserve Variability – Session II: Who Is Doing What? Mark R. Shapland, FCAS, ASA, MAAA Casualty Actuarial Society Spring Meeting San Juan, Puerto Rico.
Screening for Moral Hazard and Adverse Selection: Evidence from the Home Equity Market Discussion of Paper by Sumit Agarwal, Brent Ambrose, Souphala Chomsisengphet,
Modeling the Loss Process for Medical Malpractice Bill Faltas GE Insurance Solutions CAS Special Interest Seminar … Predictive Modeling “GLM and the Medical.
Risk and the Organization of Bank Foreign Affiliates Giovanni Dell’Ariccia IMF and CEPR Robert Marquez Arizona State University.
CIA Annual Meeting LOOKING BACK…focused on the future.
Discrete Choice Modeling William Greene Stern School of Business New York University.
Run length and the Predictability of Stock Price Reversals Juan Yao Graham Partington Max Stevenson Finance Discipline, University of Sydney.
MARKET APPRAISAL. Steps in Market Appraisal Situational Analysis and Specification of Objectives Collection of Secondary Information Conduct of Market.
1 To Loan or Not to Loan Student Coaching Notes. 2 Concepts Covered Statistics Macroeconomics Ethics.
Machine Learning 5. Parametric Methods.
Contemporary Engineering Economics, 6 th edition Park Copyright © 2016 by Pearson Education, Inc. All Rights Reserved Probabilistic Cash Flow Analysis.
1 Chapter 8: Model Inference and Averaging Presented by Hui Fang.
Personalized Recommendations using Discrete Choice Models with Inter- and Intra-Consumer Heterogeneity Moshe Ben-Akiva With Felix Becker, Mazen Danaf,
[Part 5] 1/43 Discrete Choice Modeling Ordered Choice Models Discrete Choice Modeling William Greene Stern School of Business New York University 0Introduction.
The Generalized extreme value (GEV) distribution, the implied tail index and option pricing Sheri Markose and Amadeo Alentorn Papers available at:
Discussion of Mandelbrot Themes: Alpha (Tail Index) and Scaling (H) Prepared by Sheri Markose, Amadeo Alentorn and Vikentia Provizionatou WEHIA 2005 Implications.
WELCOME TO THE PRESENTATION ON LINEAR REGRESSION ANALYSIS & CORRELATION (BI-VARIATE) ANALYSIS.
William Cook Abusing statistics in retail banks, and its contribution to the banking crisis.
Wanda Cornacchia | Banca d’Italia – Financial Stability Directorate
What is Survival Model and why it is important?
UNIT – V BUSINESS ANALYTICS
About me – Matthew Jones
Where did we stop? The Bayes decision rule guarantees an optimal classification… … But it requires the knowledge of P(ci|x) (or p(x|ci) and P(ci)) We.
Econometrics Chengyuan Yin School of Mathematics.
Stress Testing Community Banks
University of Minnesota and CEPR
RMBS Rating Methodology
Presentation transcript:

Bridging the Academic–Practitioner Divide in Credit Risk Modeling Vadim Melnitchouk, Metropoliten State University, Saint Paul, MN, US

Agenda 1. Academic model selection by a practitioner and organizational issues 2. ‘Optimal complexity model’ : stochastic parametric method with macroeconomic variables and unobserved consumer heterogeneity 3. Data access, collaboration & prototype development

Who is a practitioner? 1. Ph. D in applied math, former academic, teaching part-time ‘Data Mining’. 2. Ph. D in physics, former academic 3. M.S. in OR, former ‘Fed’ examiner 4. M.S. in Econometric

A practitioner’s search for the right academic paper /model Paper/MethodologyPotential Business ImpactOrganizational issue Andreeva, Ansell & Crook 'Modeling Profitability using Survival Combination Scores'Increase Profitability How to get CRO & CMO to agree on the same KPIs? Belloti & Crook Forecasting and Stress Testing Credit Card Default..' More accurate estimation for unexpected losses, Economic Capital Reduction US Banks are getting a stress test scenario from Regulators Fader & Hardie 'Customer- Base Analysis with Discrete- time …' Increase Sales, prevent Customer Attrition Was implemented at GE Money in Fader & Hardie 'Customer- Base Analysis with Discrete- time …'Reduce lossesCultural resistence Leow & Crook 'Intensity Models and Transition Probabilities ‘ Reduce losses Feasible, but optimal complexity model is required

Time to Default: Optimal complexity model 1. According to Bellotti & Crook (2007) survival (hazard) modeling is competitive alternative to logistic regression when predicting default events. 2. The method has become a model of choice in recent publications. But its complexity makes such technique unfeasible for practitioners. 3. It also has some limitations. Bellotti (2010) believes that ‘any credit risk model with macroeconomic variables can’t be expected to capture the direct reason for default like a loss of job, negative equity or a sudden personal crisis such as sickness or divorce’.

Methodology The goal of this paper is to present more practical method which also can take unobserved obligor heterogeneity into account. Stochastic parametric Time to Event method is well known in marketing (Hardie & Fader, 2001). It was also applied by Brusilovskiy (2005) to predict the time of the first home purchase by immigrants. The method as far as we know has not been used in credit risk by academics or practitioners.

Assumptions & inputs 1. Time to Default - Weibull distribution (Appendix) 2. Default density across obligors - Gamma distribution (to include unobserved consumer heterogeneity). 3. Vintage aggregate level modeling to avoid so called aggregation bias when unemployment is used. Inputs: 1. Monthly number of defaults 2. Time varying covariates : Unemployment and Home Price Index (HPI). Macroeconomic factors are incorporated into the hazard rate function.

Recent trends in mortgage default rate & data 1. The default rates have spiked from historical trends in 2005 and more significantly in 2006 & 2007 beginning almost immediately after origination. 2. A verage time to reach maximum default rate decreased from 5-6 (Vintage ) to 2-3 years (Vintage ) 3. LPS prime, first, fixed rate 30 years mortgage originated in 2006 data were used to build a model (Schelkle, 2011).

Model training and out-of-time validation 1. Model training period for vintage 2006 was June 2006 – March April 2009 to March 2010 period was selected for ‘out of time’ validation because unemployment increased from 8.5% to 10.1% during this period. 3. The model was implemented in MS Excel (using Solver) and in SAS/IML. Maximum likelihood was estimated to get values for five parameters.

Forecasted vs Actual monthly # of defaults Weibull/Gamma model for 2006 mortgage origination year (LPS data, vintage 2006).

Results & Discussion The forecast accuracy for ‘out-of-time’ period is at acceptable level (low forecast error and conservative estimate for regulators). Issues with one segment model: 1. Time varying covariates formula is taken from marketing application and is not flexible one for credit risk modeling (Appendix). 2. The impact of unemployment and HPI can be double counted.

Next steps in collaboration with academics 1. Bayesian parameters’ estimation was applied in collaboration with Prof. Shemyakin (St.Thomas University, St. Paul. MN) and his students to improve numeric stability. Two segments latent class Weibull model (Appendix) was also used to estimate parameters of consumer segment with default hazard increasing over time. Unemployment and HPI were not included to avoid double counting (academic’s preference).

Data access and three levels of collaboration Collaboration levelExecution by Academic's Motivation Practitioner's MotivationData Access Academic Partner Looking over your shoulderPractitioner Marketing and validation Apply new method (professional growth)N/A Prof. Fader & Prof. Hardie Joint supervisionStudent Real life project for a student Additional validation & enhancement Vintage aggregated data only Prof. Shemyakin, June 2012 Bridging the Academic– Practitioner Divide Academic and practitioner? Resolve real issue like wrong signs in multinomial regression coefficients Aggregated by delinquency status ?

Data access 1. It is very problematic to get loan level data from financial firms for joint projects. 2. Aggregate level delinquency and default data for mortgages, credit cards, installment loans and commercial lending can be extracted from public websites. 3. But data decomposition of completely aggregated data like Federal Reserve one (Appendix) should be implemented first to apply vintage based modeling.

From a prototype to production: possible collaboration Model descriptionModel CategoryScopeMajor IssuePossible solution Non-stationary Markov Chain model with hazard functions and macroeconomic variablesProduction Consumer & Commercial Zero values for some transition coefficients Bayesian estimator/ Gibbs sampling? Non-stationary Markov Chain model with multinomial transition functions and macroeconomic variablesProduction Consumer & Commercial Wrong signs in some transition coefficients? Experiment with a second order Markov ChainResearch Commercial To many parameters, small sample size for some transitions MCMC Forecasting Time to Delinquency using Stochastic Parametric ModelBenchmarkingConsumer MLE estimation numerical stability Bayesian estimation Predicting delinquent loans’ recovery using Stochastic 'Choice' ModelBenchmarkingConsumer Not included in SAS, R, etc., no standard tests Alternative to Markov model

Next search for optimal complexity model: Combined Markov Chain and Survival Analysis Model description Macroeconomic variables Objective functionMajor IssuePossible solution Leow & Crook 'Intensity Models and Transition Probabilities ‘ Next stepPartial MLE?N/A Louis, Laere, Baesens ‘Predicting bank rating transitions..’ YesPartial MLECorrelated event timesClustering Jones ‘Estimating Markov Transition’ YesLeast Sq. Migration underestimation Bayesian MCI (Christodoulakis) Kunovac ‘Estimating Credit Migration …– Bayesian Approach, NoMPLE Zero values in some transition coefficients Gibbs sampling Grimshaw & Alexander ‘Markov Chain model for delinquency..’NoMLE Statistical significance for some transitions Bayesian estimator

Conclusions Stochastic parametric method with macroeconomic variables and unobserved consumer heterogeneity can be used by practitioners as an alternative to survival modeling The optimal complexity model can provide an incentive to try to bridge the Academic –Practitioner Divide

Appendix

Latent class Weibull model with two segments Assumptions: 1. All obligors can be divided into two segments with their own fixed but unknown values of shape and scale parameters. 2. Large segment has decreasing default hazard. 3. Relatively small consumer segment exists with default hazard increasing over time. The segment size (percentage) is latent variable which must be estimated for each vintage.