Page 176 - Contributed Paper Session (CPS) - Volume 3

CPS1985 Markus Z.

Trusted official smart statistics - Challenges for
official statistics in using data sources coming
from private data producers
Markus Zwick
Federal Statistical Office of Germany

Abstract
The Federal Statistical Office of Germany engaged in the Eurostat financed
project ‘Smart business cycle statistics (SBCS)’ from the beginning of 2018 until
March 2019. The aim of the project was to use sensor data coming from
satellite and flyover observations for the production of economic indicators.
The project SBCS has shown that business cycle indicators based on satellite
data will nearly be possible in real-time and cross-border. With these
characteristics, SBCS will expand the information content fundamentally
compared to currently used business cycle indicators. In general, National
Statistical Institutes (NSIs) will be able to produce SBCS in the future, but
probably with external partners and sometimes even in cooperation with
private data producers. The paper discusses the production of official statistics
in the IoT age, on the example and the experience made at the project SBCS.
IoT based official statistical products will probably mostly be work-sharing
products, with specialists in the three working fields ‘conception and quality’,
‘data engineering’ and ‘data visualization’. For getting high quality statistical
products, the rules for integrating private data products into Official Statistics
have to be developed.

Keywords
New digital data sources; Official Statistics; Privately held data; Quality; Skills

1. Introduction
The opportunities and challenges National Statistical Offices (NSIs) are
facing are growing with the digital revolution. In general, official statistics will
be faster, more precise for small groups and all of that with fewer burdens for
the respondents in the future. Therefore, statistical offices have to be
transformed in a modern information provider. Otherwise private data
producers will do more of what we call official statistics.
The paper will describe some of the challenges of the digital
transformation process for statistical offices. Big Data, meanwhile called as
‘new digital data’ or ‘non-traditional data’ allow a deeper and more precise
picture of the society and the economy. Mobile phone, satellite as well as data
from the internet add information to official statistical products, in the
meanwhile and even more in the future. In some cases, these kinds of new

165 | I S I W S C 2 0 1 9