Iain Brown Ph.D.

Iain Brown Ph.D.

Marlow, England, United Kingdom
35K followers 500+ connections

Join to follow

About

Dr. Iain Brown is the Head of Data Science for Northern Europe and Adjunct Professor of…

Articles by Iain

See all articles

Contributions

Join now to see all contributions

Activity

Join now to see all activity

Experience

  • SAS Graphic

    SAS

    Marlow, United Kingdom

  • -

    Southampton, United Kingdom

  • -

    Marlow, United Kingdom

  • -

    Marlow, United Kingdom

  • -

    London, United Kingdom

  • -

    Southampton, United Kingdom

  • -

    London, United Kingdom

Education

  • University of Southampton Graphic

    University of Southampton

    -

    The purpose of this Doctoral research was to determine the most appropriate machine learning techniques for modelling the three key components of the Basel II minimum capital requirement: probability of default (PD), loss given default (LGD), and exposure at default (EAD).

    Throughout my research I developed a variety of novel modelling approaches to solve real world issues relating to financial institutions implementing the Basel II capital requirements. I presented my research at a…

    The purpose of this Doctoral research was to determine the most appropriate machine learning techniques for modelling the three key components of the Basel II minimum capital requirement: probability of default (PD), loss given default (LGD), and exposure at default (EAD).

    Throughout my research I developed a variety of novel modelling approaches to solve real world issues relating to financial institutions implementing the Basel II capital requirements. I presented my research at a number of internationally renowned conferences and have subsequently had three unique pieces of work published.

  • -

    - Applied Statistics, Computer Modelling, Simulation and Mathematical Programming, Model building in Mathematical Programming, Integer Programming and Advanced Techniques in OR

  • -

    - Quantitative Statistical Methods, Applied Statistical Modelling, Linear Algebra, Econometrics, Computer Modelling for Operational Research, Techniques and Methods in Operational Research and Marketing Analytics

  • -

Licenses & Certifications

Publications

  • Mastering Marketing Data Science: A Comprehensive Guide for Today's Marketers

    Wiley

    Unlock the Power of Data: Transform Your Marketing Strategies with Data Science

    In the digital age, understanding the symbiosis between marketing and data science is not just an advantage; it's a necessity. In Mastering Marketing Data Science: A Comprehensive Guide for Today's Marketers, Dr. Iain Brown, a leading expert in data science and marketing analytics, offers a comprehensive journey through the cutting-edge methodologies and applications that are defining the future of marketing.…

    Unlock the Power of Data: Transform Your Marketing Strategies with Data Science

    In the digital age, understanding the symbiosis between marketing and data science is not just an advantage; it's a necessity. In Mastering Marketing Data Science: A Comprehensive Guide for Today's Marketers, Dr. Iain Brown, a leading expert in data science and marketing analytics, offers a comprehensive journey through the cutting-edge methodologies and applications that are defining the future of marketing. This book bridges the gap between theoretical data science concepts and their practical applications in marketing, providing readers with the tools and insights needed to elevate their strategies in a data-driven world. Whether you're a master's student, a marketing professional, or a data scientist keen on applying your skills in a marketing context, this guide will empower you with a deep understanding of marketing data science principles and the competence to apply these principles effectively.

    - Comprehensive Coverage: From data collection to predictive analytics, NLP, and beyond, explore every facet of marketing data science.
    - Practical Applications: Engage with real-world examples, hands-on exercises in both Python & SAS, and actionable insights to apply in your marketing campaigns.
    - Expert Guidance: Benefit from Dr. Iain Brown's decade of experience as he shares cutting-edge techniques and ethical considerations in marketing data science.
    - Future-Ready Skills: Learn about the latest advancements, including generative AI, to stay ahead in the rapidly evolving marketing landscape.
    - Accessible Learning: Tailored for both beginners and seasoned professionals, this book ensures a smooth learning curve with a clear, engaging narrative.

    Mastering Marketing Data Science is designed as a comprehensive how-to guide, weaving together theory and practice to offer a dynamic, workbook-style learning experience.

    See publication

  • Machine learning interpretability for a stress scenario generation in credit scoring based on counterfactuals

    Expert Systems with Applications

    To boost the application of machine learning (ML) techniques for credit scoring models, the blackbox problem should be addressed. The primary aim of this paper is to propose a measure based on counterfactuals to evaluate the interpretability of a ML credit scoring technique. Counterfactuals assist with understanding the model with regard to the classification decision boundaries and evaluate model robustness. The second contribution is the development of a data perturbation technique to…

    To boost the application of machine learning (ML) techniques for credit scoring models, the blackbox problem should be addressed. The primary aim of this paper is to propose a measure based on counterfactuals to evaluate the interpretability of a ML credit scoring technique. Counterfactuals assist with understanding the model with regard to the classification decision boundaries and evaluate model robustness. The second contribution is the development of a data perturbation technique to generate a stress scenario.

    We apply these two proposals to a dataset on UK unsecured personal loans to compare logistic regression and stochastic gradient boosting (SBG). We show that training a blackbox model (SGB) as conditioned on our data perturbation technique can provide insight into model performance under stressed scenarios. The empirical results show that our interpretability measure is able to capture the classification decision boundary, unlike AUC and the classification accuracy widely used in the banking sector.

    See publication

  • Exposure at default models with and without the credit conversion factor

    European Journal of Operational Research

    The Basel II and III Accords allow banks to calculate regulatory capital using their own internally developed models under the advanced internal ratings-based approach (AIRB). The Exposure at Default (EAD) is a core parameter modelled for revolving credit facilities with variable exposure. The credit conversion factor (CCF), the proportion of the current undrawn amount that will be drawn down at time of default, is used to calculate the EAD and poses modelling challenges with its bimodal…

    The Basel II and III Accords allow banks to calculate regulatory capital using their own internally developed models under the advanced internal ratings-based approach (AIRB). The Exposure at Default (EAD) is a core parameter modelled for revolving credit facilities with variable exposure. The credit conversion factor (CCF), the proportion of the current undrawn amount that will be drawn down at time of default, is used to calculate the EAD and poses modelling challenges with its bimodal distribution bounded between zero and one. There has been debate on the suitability of the CCF for EAD modelling. We explore alternative EAD models which ignore the CCF formulation and target the EAD distribution directly. We propose a mixture model with the zero-adjusted gamma distribution and compare its performance to three variants of CCF models and a utilization change model which are used in industry and academia. Additionally, we assess credit usage – the percentage of the committed amount that has been currently drawn – as a segmentation criterion to combine direct EAD and CCF models. The models are applied to a dataset from a credit card portfolio of a UK bank. The performance of these models is compared using cross-validation on a series of measures. We find the zero-adjusted gamma model to be more accurate in calibration than the benchmark models and that segmented approaches offer further performance improvements. These results indicate direct EAD models without the CCF formulation can be an alternative to CCF based models or that both can be combined.

    Other authors
    See publication

  • Developing Credit Risk Models Using SAS Enterprise Miner and SAS/STAT: Theory and Applications

    SAS Press

    Combine complex concepts facing the financial sector with the software toolsets available to analysts.

    The credit decisions you make are dependent on the data, models, and tools that you use to determine them. Developing Credit Risk Models Using SAS Enterprise Miner and SAS/STAT: Theory and Applications combines both theoretical explanation and practical applications to define as well as demonstrate how you can build credit risk models using SAS Enterprise Miner and SAS/STAT and apply…

    Combine complex concepts facing the financial sector with the software toolsets available to analysts.

    The credit decisions you make are dependent on the data, models, and tools that you use to determine them. Developing Credit Risk Models Using SAS Enterprise Miner and SAS/STAT: Theory and Applications combines both theoretical explanation and practical applications to define as well as demonstrate how you can build credit risk models using SAS Enterprise Miner and SAS/STAT and apply them into practice.

    The ultimate goal of credit risk is to reduce losses through better and more reliable credit decisions that can be developed and deployed quickly. In this example-driven book, Dr. Brown breaks down the required modeling steps and details how this would be achieved through the implementation of SAS Enterprise Miner and SAS/STAT.

    Users will solve real-world risk problems as well as comprehensively walk through model development while addressing key concepts in credit risk modeling. The book is aimed at credit risk analysts in retail banking, but its applications apply to risk modeling outside of the retail banking sphere. Those who would benefit from this book include credit risk analysts and managers alike, as well as analysts working in fraud, Basel compliancy, and marketing analytics. It is targeted for intermediate users with a specific business focus and some programming background is required.

    Efficient and effective management of the entire credit risk model lifecycle process enables you to make better credit decisions. Developing Credit Risk Models Using SAS Enterprise Miner and SAS/STAT: Theory and Applications demonstrates how practitioners can more accurately develop credit risk models as well as implement them in a timely fashion.

    See publication

  • An experimental comparison of classification algorithms for imbalanced credit scoring data sets

    Expert Systems with Applications

    In this paper, we set out to compare several techniques that can be used in the analysis of imbalanced credit scoring data sets. In a credit scoring context, imbalanced data sets frequently occur as the number of defaulting loans in a portfolio is usually much lower than the number of observations that do not default. As well as using traditional classification techniques such as logistic regression, neural networks and decision trees, this paper will also explore the suitability of gradient…

    In this paper, we set out to compare several techniques that can be used in the analysis of imbalanced credit scoring data sets. In a credit scoring context, imbalanced data sets frequently occur as the number of defaulting loans in a portfolio is usually much lower than the number of observations that do not default. As well as using traditional classification techniques such as logistic regression, neural networks and decision trees, this paper will also explore the suitability of gradient boosting, least square support vector machines and random forests for loan default prediction.

    Five real-world credit scoring data sets are used to build classifiers and test their performance. In our experiments, we progressively increase class imbalance in each of these data sets by randomly under-sampling the minority class of defaulters, so as to identify to what extent the predictive power of the respective techniques is adversely affected. The performance criterion chosen to measure this effect is the area under the receiver operating characteristic curve (AUC); Friedman’s statistic and Nemenyi post hoc tests are used to test for significance of AUC differences between techniques.

    The results from this empirical study indicate that the random forest and gradient boosting classifiers perform very well in a credit scoring context and are able to cope comparatively well with pronounced class imbalances in these data sets. We also found that, when faced with a large class imbalance, the C4.5 decision tree algorithm, quadratic discriminant analysis and k-nearest neighbours perform significantly worse than the best performing classifiers.

    Other authors
    • Christophe Mues
    See publication

  • Benchmarking regression algorithms for loss given default modeling

    International Journal of Forecasting

    The introduction of the Basel II Accord has had a huge impact on financial institutions, allowing them to build credit risk models for three key risk parameters: PD (probability of default), LGD (loss given default) and EAD (exposure at default). Until recently, credit risk research has focused largely on the estimation and validation of the PD parameter, and much less on LGD modeling. In this first large-scale LGD benchmarking study, various regression techniques for modeling and predicting…

    The introduction of the Basel II Accord has had a huge impact on financial institutions, allowing them to build credit risk models for three key risk parameters: PD (probability of default), LGD (loss given default) and EAD (exposure at default). Until recently, credit risk research has focused largely on the estimation and validation of the PD parameter, and much less on LGD modeling. In this first large-scale LGD benchmarking study, various regression techniques for modeling and predicting LGD are investigated. These include one-stage models, such as those built by ordinary least squares regression, beta regression, robust regression, ridge regression, regression splines, neural networks, support vector machines and regression trees, as well as two-stage models which combine multiple techniques. A total of 24 techniques are compared using six real-life loss datasets from major international banks. It is found that much of the variance in LGD remains unexplained, as the average prediction performance of the models in terms of R2 ranges from 4% to 43%. Nonetheless, there is a clear trend that non-linear techniques, and in particular support vector machines and neural networks, perform significantly better than more traditional linear techniques. Also, two-stage models built by a combination of linear and non-linear techniques are shown to have a similarly good predictive power, with the added advantage of having a comprehensible linear model component.

    Other authors
    See publication

  • Regression Model Development for Credit Card Exposure At Default (EAD) using SAS/STAT and SAS Enterprise Miner

    SAS Global Forum Conference Proceedings

    In this paper, we propose a comprehensive and robust model for predicting the exposure at default (EAD). For off-balance sheets (for example, credit cards) to calculate the EAD, one requires the committed but unused loan amount times a credit conversion factor (CCF). Ordinary least squares, logistic and cumulative logistic regression models are analyzed with the primary aim of finding the most robust and comprehensible model for the prediction of the CCF.

    A real-life data set with…

    In this paper, we propose a comprehensive and robust model for predicting the exposure at default (EAD). For off-balance sheets (for example, credit cards) to calculate the EAD, one requires the committed but unused loan amount times a credit conversion factor (CCF). Ordinary least squares, logistic and cumulative logistic regression models are analyzed with the primary aim of finding the most robust and comprehensible model for the prediction of the CCF.

    A real-life data set with monthly balance amounts for clients over the period 2001–2004 is used in the building and testing of the regression models. Parameter estimates and comparative statistics are then given to determine the best overall model.

    See publication

Honors & Awards

  • Associate Fellow of the Operational Research Society (AFORS)

    The OR Society

    The Associate Fellow of The OR Society (AFORS) accreditation is awarded to those with a successful track record in O.R. extending over at least five years. Admission to the category of Associate Fellow indicates a successful record of achievement in, and/or a significant contribution to, operational research. Candidates must have a minimum of five years’ experience in O.R., which could include periods in masters’ level training, practice, research or education, and normally hold at least a…

    The Associate Fellow of The OR Society (AFORS) accreditation is awarded to those with a successful track record in O.R. extending over at least five years. Admission to the category of Associate Fellow indicates a successful record of achievement in, and/or a significant contribution to, operational research. Candidates must have a minimum of five years’ experience in O.R., which could include periods in masters’ level training, practice, research or education, and normally hold at least a second class honours degree (or equivalent qualification).

  • SAS Student Ambassador 2011

    SAS Institute

    The SAS Student Ambassador Program is a competitive program that recognises and supports students who use SAS technologies in innovative ways that benefit their respective industries and fields of study. Select students will be named SAS Student Ambassadors and earn the opportunity to present their research at the yearly SAS Global Forum in the USA.

    http://support.sas.com/learn/ap/student/amb.html

Recommendations received

1 person has recommended Iain

Join now to view

More activity by Iain

View Iain’s full profile

  • See who you know in common
  • Get introduced
  • Contact Iain directly
Join to view full profile

Other similar profiles

Explore more posts

Explore collaborative articles

We’re unlocking community knowledge in a new way. Experts add insights directly into each article, started with the help of AI.

Explore More

Add new skills with these courses

See all courses