Written by Manfred Hübner | 02 November 2026
Posted in Database news

As part of a comprehensive quality review of our database, we have fully recalculated the history of the sentix Risk Aversion Index and of the risk appetite series in the sentix Styles Index. The series are now complete and follow consistently from the underlying survey results. As a result, a number of historical values have changed, mostly only marginally.

What has changed

• A missing month added. The derived figures for the December 2018 survey were previously not available. They have now been added, so the series run without interruption back to February 2004.
• Corrections to individual survey results carried through. On a few dates since December 2019, corrected results of individual survey questions had not been reflected in the figures derived from them. We have recalculated these figures. On the dates concerned, the changes are more clearly visible.
• Minor knock-on effect on later months. The standardised values of the Risk Aversion Index are calculated each month relative to the entire history. When an earlier value changes, the values of subsequent months shift slightly as well, as a rule only in the fourth decimal place.

Which series are affected

• sentix Risk Aversion Index: Equities, Bonds, Overall Index, Individuals vs. Institutionals and Equities vs. Bonds, from December 2019 or July 2022
• sentix Styles Index: Risk Appetite (Equities, Bonds, Overall Index) on individual dates
• Emerging Markets Asset Allocation Bias: only the addition of December 2018

All other sentix series and all values before December 2018 remain unchanged.

Our recommendation

If you use these series in your own models or analyses, please reload the history from December 2018 onwards. The updated values are available in the charts, in Bloomberg, the data download, the DataLibrary and via the REST API.

This review is part of our ongoing work on the quality and traceability of sentix data. Please do not hesitate to contact us if you have any questions.