Credit risk is always s hot topic in economic news. Are US Treasury bonds safe? Does the Chinese banking system have problems? Is the ratio of household debt to GDP is high in Malaysia? etc. When we think about credit risk, it is critically important to know how we measure credit risk. So I would like to reconsider how we can apply Machine learning to credit risk management in financial institutions. When I used to be a credit risk manager in a Japanese consumer finance company 10 years ago, there were no cloud service and no social media. However, now that it is the time of big data, it is a good timing to reconsider how we can manage credit risk by using the latest technologies such as cloud services and Machine learning. Let me take three points as follows.
1. Statistical models
One of the key metrics in credit risk management is the probability of defaults (PD). It is usually calculated from statistical models such as regression analysis. Machine learning has algorithms about regression analysis, therefore it must be easy to implement Machine learning to credit risk management systems in financial intuitions. Once PD is calculated from Machine learning, this figure can be used the same way as current practices in financial institutions. Statistical models usually have versions since it is developed but there is no need to worry about version control because it is easy on the cloud system.
Data is more different from one which used to be 10 years ago. I used private and closed data only in order to calculate the risk for each borrower. Now that social media data and open public data are available in risk management, we need to consider how to use these kind of data in practice. For this purpose, cloud services are good because they are scalable and easy to expand the capacity to store data whenever it is needed.
3. Product development
Recently product development is fast and active to keep the competitive edge as customer can choose channels, to contact with financial institutions and are getting more demanding. That is why risk management in financial institutions should be flexible to update product portfolio and adjust methods in the light of characteristics of new products. Combination of cloud service and machine learning enable us to develop the risk models, enough to quick to keep up with new products and change of business environments.
Unlike retail industries, financial industries are heavily regulated and required to audit risk management systems periodically. Therefore, audit trails are also important when Machine learning is applied to credit risk management. I think combination cloud services and machine learning is good to enhance credit risk management with cost-effectiveness in the long run. I would like to try this combination in credit risk management if I can do it.