Using AI to reduce subjectivity in credit analysis
Developing custom AI tools to mitigate analyst biases
We have joined forces with Chenavari Financial Group to develop custom AI tools aimed at improving the quality and mitigating the subjectivity within credit analysis. The project aims to enhance transparency, trustworthiness, and adoption through cutting-edge AI technologies.
By harnessing the wealth of existing data and employing a range of AI techniques, the project aims to enhance efficiency, pricing capability, and precision in forecasting within credit analysis.
The collaboration
We are working with Chenavari to understand the unique challenges facing the credit industry such as factor analysis, attribution and pricing of complex instruments such as CLOs (collateralised loan obligations).
The project aims to incorporate cutting edge AI technologies to solve these challenges in real-world situations in a way that enhances transparency, trustworthiness and widespread adoption.
Achieving this goal will be a collaborative effort to utilise the vast data available to Chenavari, and interpreting it using an array of cutting edge AI techniques. Some key aspects we will be looking to optimise include efficiency, pricing capability and more precise forecasting within credit analysis.
Government & Academic support
The initiative is supported by UK Research and Innovation (UKRI) and administered by Innovate UK, a government sponsorship body that promotes industry innovation in key sectors.
The project will be supported by Dr John Cartlidge at the University of Bristol, providing AI and financial modelling expertise, and Prof. Karen Elliott at the University of Birmingham, an expert in the trustworthy application of AI.
Anticipated impact
We anticipate that this project will bring significant advantages to the credit industry. The ultimate outcome of the project is to extend beyond its initial credit-focused scope, generating positive effects on the broader economy.
We are delighted to be working with one of the leading and most innovative structured credit firms. This gives us the opportunity to showcase the full potential of our Stratlib.AI technology in a real-world setting and I’m particularly excited about the benefits we can bring directly to Chenavari and their clients.
Daniel Gold, CEO, Stratiphy
Chenavari believes there is an opportunity to use AI technology to reshape the way credit fundamental analysis is performed. In the past algorithms have tended to focus on technical signals and require a huge amount of data in a few dimensions only. Credit fundamental analysis tends to be done by specialists who analyse a lot of dimensions but focus on specific sectors and geographies. AI enables us to consider data of all shapes and sizes and to create tools to improve the speed and objectivity of analysis.
Ed Smalley, Chief Risk Officer, Chenavari
Q&A with Ed and Daniel
Ed Smalley, Chief Risk Officer, Chenavari
What problem are you looking to solve, and why is AI the right tool for the job?
There is a great need for standardisation within the credit analysis world. Inconsistencies between analysts can lead to poor outcomes and a differentiated client experience.
Furthermore, it is challenging to incorporate vast datasets, look for hidden connections and efficiently forecast, or take a view, on key parameters. These challenges impede the quality of credit analysis.
The time is right to use AI to solve these real-world problems. The power of AI has increased considerably of late and its availability as a tool to improve outcomes has been embraced across many industries, so it is time we did the same in credit analysis.
Why is this important to Chenavari?
The project will enable us to process more data, more efficiently, and improve the performance of our pricing analysis. At the same time, it will help to eliminate inconsistencies due to human factors like bias and preferences in the market.
What opportunities does this create for Chenavari in the credit markets?
By utilising this technology, we have the opportunity to be pioneers in the space by innovating and being more reactive to changing markets. In turn, this will help us to identify new opportunities and offer better value to clients.
Are there any risks in looking to automate the work you currently do?
We are not going to replace the analytical work we currently undertake, this project is aimed at supplementing it. It will make the job for our analysts easier by incorporating vast datasets in real time that they currently cannot do and will improve the consistency of our analytics across the board.
Why did you choose to partner with Stratiphy on this project?
They are a highly innovative Fintech company and their approach can greatly enhance the work we do at Chenavari. By combining this approach with our Credit Analysis capabilities, we believe the result can provide a significant advantage yielding great benefits for each client.
Daniel Gold, CEO, Stratiphy
What role does transparency and trustworthiness play in the development of these AI-powered tools?
We don’t believe in a black-box approach to financial analysis, the results are best used when they are explained and understood. Trustworthiness and transparency are therefore key to the adoption of AI by end-users.
Stratiphy's collaborative effort with Chenavari aims to bring a new level of credibility to credit-related securities pricing. Our approach to the integration of AI ensures a transparent and reliable approach to credit analysis by focussing on the outputs being comprehensible and empowering users with control.
Why do you think the government is playing such an active role in promoting these developments in AI?
The government has identified that the UK is a leading centre of AI talent but falls short when it comes to implementing and adopting this technology in financial services. Their intention is to promote a trustworthy implementation of AI in order to improve the UK’s competitive edge, and open new opportunities for UK businesses. Ultimately, they want the UK to be a leader in AI and are putting funding behind this project to enable that.
Why is this important for Stratiphy?
We have developed a highly flexible toolkit for financial analysis that applies to a wide range of scenarios within financial time series analysis. This includes portfolio management, signal generation and credit analysis. Developing this capability with AI is a natural progression and working with Chenavari as a leader in the credit field means we can provide more benefit to the market.
Are there any ethical considerations regarding the project? How are these going to be dealt with?
We take ethical considerations of AI seriously, and that is why we have teamed up with Prof Karen Eliottt who is an authority in this space. This project focussed on identifying ethical issues, biases or spurious outputs and mitigating these as risks.
What AI tools will the project use?
This project involves the use of a variety of AI techniques from optimisation algorithms to image processing and Natural Language Processing (NLP). We are excited by the interdisciplinary nature of the problem and the opportunity to apply these techniques within a new field.