Abstract

Crises can challenge decision making, and many of the decisions made during crises can have a huge impact. Therefore, especially in these situations, it is important to have as reliable information as possible to support these decisions. Statistics Finland (SF) has a comprehensive data lake from Finnish registers and other data sources. Some of this data is also very timely, for example, the Income register includes every citizen's income information with a one-day delay, and changes in this register are transferred to SF every night. To support data-driven decision making, SF, VATT Institute for Economic Research (VATT), and Helsinki Graduate School of Economics (HGSE) established a Data Room to make fast analyses for decision makers. A key element for fast data analysis is having the competence to deliver relevant data quickly to researchers. The solution for this is a so-called ready-made research data library. In this library, data is compiled into small modules that are updated regularly (e.g., weekly, monthly, or yearly), and the data is pseudonymized.
Because data subjects can indirectly be identified from the confidential microdata, data is only accessible via the remote access system FIONA, which Statistics Finland developed together with the state-owned company IT Center for Science. This system offers a secure way to handle sensitive data. FIONA is a closed environment without an external internet connection, and researchers have no possibility to take out or upload any data or any files into the system. In order to access the data, researchers must also apply for a user license. Every application is evaluated separately, and researchers are granted access only to those datasets they need for their current research.
In this presentation, I also provide some use cases that show that data-driven decision making can be highly effective.