A direct correlation exists between data quality and productivity improvements within the risk management function. Poor quality data can result in increased time to develop models, lower confidence in the model results and less time to analyze results. Less precise modelling caused by poor data quality can mean that banks have to set aside higher capital buffers and loss allowance provisions.
There are numerous processes available for banks to define data quality, and guiding principles that can be implemented to improve data quality. When defining the firm’s framework for data quality in risk analytics, the following guidelines can be applied:
- Specifically document how data are defined and constructed.
- Ensure that data accurately quantifies the concept that modellers intend to measure.
- Independently verify numerical correctness by using backlinks to primary sources, quality declarations, unique identifiers and accessible quality logs.
Moody’s Analytics recently published an article entitled “When Good Data Happen to Good People: Boosting Productivity with High-Quality Data.” This article quantifies the impact of data quality on improvements to analytical productivity, and provides a functional definition of data quality along with detailed examples of the impact of improving data quality on efficiency in analytical tasks. To read the full whitepaper, click here.