Robust index design

Key takeaways:

• Index sensitivity to data errors varies among data inputs, in particular sustainability data
• Different index construction methodologies exhibit different sensitivity to data errors, underscoring the importance of methodology selection
• Understanding the sensitivity to data errors is key to robust index construction

Points of differentiation:

• We propose an original and efficient method of input-sensitivity analysis
• We introduce a new and efficient methodology for assessing how index performance responds to variations in input data
• The results are intuitive and expressed in terms of turnover/costs to correct data errors
• Inclusion of data-sensitivity analysis facilitates enhanced index design

What does our research mean for investors?

• We examine our index design to strike a balance between specificity to the signal vs. robustness against noise. This approach helps support more stable, transparent, and investable benchmarks that align with institutional performance and risk objectives
• With this tool, choosing the most suitable methodology can be straightforward and intuitive for investors