Welcome to the Paragon resources hub
Here you’ll find a collection of useful materials on the techniques and considerations when it comes to developing, implementing, using and managing credit risk models, as well as the latest Paragon opinion and industry developments.
What do credit risk modelling and F1 have in common?
A light-hearted look at how F1 factors such as fuel mix, tyres, cockpit information, reliability, driver skills and safety compare with assessing whether to use machine learning or traditional logistic regression models in credit risk.
In a world of machine learning, is champion / challenger decision testing dead?
How do we ensure the better scores available from machine learning models become better decisions, when actions change the outcomes you are trying to predict? Mark Thompson argues that how you use machine learning models and test their outcomes is critically important.
Data selection bias - can it be a matter of life or death?
Data gets used to create models, that get used to make decisions, such as whether an application is accepted or rejected, how much someone can borrow, for how long and at what price. Or what to sell, which transactions to authorise, whether to increase or decrease limits, how and when to chase payments and more.
With bad decisions having a profound effect on lender and customer, how does data selection bias impact the modelling that drives those decisions?
Paragon: the credit expert’s choice
“We get value from Modeller due to its efficiency, time saving and auditing features.”
CARLIEN KRUGER, SENIOR MODELLER, WESTPAC
Find out why credit risk analytics experts choose Paragon software.
Our software
Whether you’re building and deploying models, automating decisions or managing model risk and governance, Paragon’s software comes with our no compromise, valued engineering built in.