1 / 25

Creating synthetic sub-regional baseline populations

Creating synthetic sub-regional baseline populations. Dr Paul Williamson Dept. of Geography University of Liverpool. Collaborators: Robert Tanton (NATSEM, Australia) Ludi Simpson (CCSR, UK) Maja Zaloznik (Liverpool, UK). Local area microdata containing local-area distributions

Télécharger la présentation

Creating synthetic sub-regional baseline populations

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Creating synthetic sub-regional baseline populations Dr Paul Williamson Dept. of Geography University of Liverpool Collaborators: Robert Tanton (NATSEM, Australia) Ludi Simpson (CCSR, UK) Maja Zaloznik (Liverpool, UK)

  2. Local area microdata containing local-area distributions [eg. smoking by income by sub-region] 1. Context • What do we want?? b) What have we got?

  3. nd SAR District: Leeds (2 largest in UK) 95% Count % Confidence Economic position Female Total female Interval Employee full - time 1525 4146 36.8 ±1.5 On a Govt scheme 31 77 40.3 ±11.0 Unemployed 168 573 29.3 ±3.7 Retired 12 67 2116 59.9 ±2.1 52.9 Total 5545 10485 ±1.0 Over-exaggerate problem? • Large-scale survey • 2% sample • Minimally multivariate • Not based on minorities (e.g. unemployed ethnic minority) • Min. geog. threshold: 120k • Decadal

  4. Local income distribution Survey distribution [smoking x income] Local smoking distribution Solution Reweight survey data... ...BUT weighting DOWN instead of up  Synthetic microdata

  5. 2. IPF (Raking) Understanding IPF… N.B. IPF = Raking = IPF Q.What is IPF/Raking doing? A. Preserving the Odds ratios ...

  6. CAVEAT: variation independence

  7. 3. Combinatorial Optimisation Male Female TARGET: 5 5 Young 2 Old 8 • Guided incremental weight adjustment

  8. 4. IPF/Raking v CO

  9. Comparison for margin-constrained tables • Target: age x sex x tenure x economic position (64 counts) at district level (17 districts) % NFC (17 district average) 32 37 22 18 • 2% SAR • IPFU • IPFN • CO

  10. Simpson & Tranmer (2005) • Target: Car ownership (2) x Tenure (3) (6 counts; 3%s) for residents at ward level

  11. 5. GREGWT

  12. Understanding GREGWT…

  13. 6. GREGWT v CO

  14. Fit toconstraint variables (74 counts): GREGWT ‘convergent’ SLAs in NSW: Fit toconstraint variables (74 counts): GREGWT ‘NON-convergent’ SLAs in NSW

  15. Fit tomargin-constrained distribution (household income x mortgage/rent): GREGWT ‘convergent’ SLAs

  16. 7. Variation idependence (again...)

  17. UNIVARIATEconstraints (158 constrained counts)

  18. BIVARIATEconstraints (586 constrained counts)

  19. Local socio-economics Survey data [District-level socio-demographics] Estimated GP Patient socio-economic characteristics Estimated HE Student socio-economic characteristics GP Patient age, sex, location HE Student age, sex, location 8. Conclusion (a) Accuracy of estimates (fitness for purpose?) (b) Unanswered questions (c) Applications in the real world…

More Related