Abstract
AbstractIn this paper we consider estimation of unobserved components in state space models using a dynamic factor approach to incorporate auxiliary information from high-dimensional data sources. We apply the methodology to unemployment estimation as done by Statistics Netherlands, who uses a multivariate state space model to produce monthly figures for unemployment using series observed with the labour force survey (LFS). We extend the model by including auxiliary series of Google Trends about job-search and economic uncertainty, and claimant counts, partially observed at higher frequencies. Our factor model allows for nowcasting the variable of interest, providing reliable unemployment estimates in real-time before LFS data become available.
Highlights
In this paper we investigate how “Big Data” can be incorporated into estimation of unobserved components using state space models
In line with the aforementioned paper, we extend the state space model used by Statistics Netherlands in order to combine the survey data with a high-dimensional auxiliary series, as it could yield more information a univariate one, which is not affected by publication lags and that can eventually be observed at a higher frequency than the labour force series
This paper proposes a method to include a high-dimensional auxiliary series in a state space model in order to improve the estimation and nowcast of unobserved components
Summary
In this paper we investigate how “Big Data” can be incorporated into estimation of unobserved components using state space models. These figures are, considered too volatile to produce sufficiently reliable monthly estimates for the employed and the unemployed labour force at monthly frequency For this reason Statistics Netherlands estimates monthly unemployment figures, together with its change, as unobserved components in a state space model where the observed series come from the monthly Dutch LFS, using a model originally proposed by Pfeffermann (1991). In line with the aforementioned paper, we extend the state space model used by Statistics Netherlands in order to combine the survey data with a high-dimensional auxiliary series, as it could yield more information a univariate one, which is not affected by publication lags and that can eventually be observed at a higher frequency than the labour force series.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: Journal of the Royal Statistical Society Series A: Statistics in Society
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
