Rocky Mountain Research Station Publications

RMRS Online Publication - Journal Articles, External Publications, and Special Reports
Combining survey and administrative data using state space models

Horn, Stephen; Czaplewski, Ray. 2013. Combining survey and administrative data using state space models. In: Proceedings: NTTS - Conference on New Techniques and Technologies for Statistics; Brussels; 5-7 March 2013. Eurostat. doi: 10.2901/Eurostat.C2013.001

Even as access to transactional data has been transformed by harnessing electronic flows, use of satellite imagery, research access to linked customer level records, and harmonising collections across jurisdictions, official statisticians are under pressure to detect significant turning points within response times and resolutions that cannot be handled by present estimation methods.

State space models can be used to combine sources of data efficiently while respecting quality demands in advising government decision making. Specifically we phrase the measure problem as how to combine high quality, high cost, unit level information obtained from a sparse sample with 'short' granular population views in an optimal manner with calculable error structure. We indicate the potential of recursive predictive methods to deliver satisfactory estimates from repeated surveys, and from systems for monitoring public forest cover and for welfare payment assurance.

Keywords: Kalman filters, measurement error, restriction estimators

About PDFs: For best results, do not open the PDF in your Web browser. Right-click on the PDF link to download the PDF file directly to your computer. Click here for more PDF help.

Download Article

PDF File Size: 85 K

Title: RMRS Other Publications: Combining survey and administrative data using state space models
Electronic Publish Date: April 2, 2013
Last Update:
April 2, 2013

RMRS Publications | Order a publication | Contact Us