Online Program

Return to main conference page

All Times ET

Program is Subject to Change

Wednesday, June 16
Wed, Jun 16, 1:30 PM - 3:30 PM
TBD
Minimizing Revisions for a Monthly Indicator

Nowcasting Versus Advanced Surveys for Estimating Austrian Principal Economic Indicators (308073)

*Markus Froehlich, Statistics Austria 

Keywords: robust time series, advanced surveying, multivariate time series, imputation

Currently Statistics Austria provides principal economic indicators (index of production, turnover index, etc.) 55 days after the end of the reference period. With the European Framework Regulation Integrating Business Statistics (FRIBS) which will be implemented in 2023 the submission deadline will be shortened to 40 days after the end of the reference month. At this time about 40 per cent of reponses are missing. Different estimation strategies have been tested in order to obtain early estimates: extrapolation of univariate time series for key variables which are used for the imputation of related variables (used as a benchmark method), multivariate time series models on an aggregated level integrating information of early respondents as well as advanced surveys based on early respondents. This paper describes the methodology for generating early estimates with both, multivariate time series methods and advanced sampling methods followed by a discussion of the results achieved with the individual estimation methods.