The Uncertainty Perception Indicator developed by DoCMA uses data of media coverage to quantify perceived uncertainty and its drivers. The German version consists of the business daily "Handelsblatt", the center-left daily "Süddeutsche Zeitung" and the center-right daily "Die Welt".
The current UPI version uses the RollingLDA method to identify the sources of economic uncertainty. Researchers are welcome to use our data for their own analyses (Please cite as: Rieger, J., Hornig, N., Schmidt, T. and Müller, H. (2023). Early Warning Systems? Building Time Consistent Perception Indicators for Economic Uncertainty and Inflation Using Efficient Dynamic Modeling. Accepted for MUFin’23.) The data of the latest UPI version is free for download. The data set includes:
- The UPI values till March 2023 (Analysis corpus relative to entire corpus, monthly data)
- The UPI LDA's topic shares over time (word occurences relative to entire corpus, monthly data)
- The UPI LDA's topic words over time (absolute occurene, monthly data)
For question on methodology see the respective working paper.