Small Area Estimation Programme

Slides:



Advertisements
Similar presentations
1 Combining migration data from multiple sources: Applications to internal movements in England, James Raymer with Peter W.F. Smith and Corrado.
Advertisements

1 Philip Clarke and Denise Silva Development of Small Area Estimation at ONS.
Riku Salonen Regression composite estimation for the Finnish LFS from a practical perspective.
A model-based approach for estimating international emigration for local authorities Brian Foley, Office for National Statistics BSPS day meeting London.
Small Area Estimates of Fuel Poverty in Scotland Phil Clarke (ONS), Ganka Mueller (Scottish Government)
1 Case Study 1: How to Deal with Estimates with Low Reliability 2009 Population Association of America ACS Workshop April 29, 2009.
Access to UK Census Data for Spatial Analysis: Towards an Integrated Census Support Service John Stillwell 1, Justin Hayes 2, Rob Dymond-Green 2, James.
Beyond 2011 – A new paradigm for population statistics? Pete Benton, Beyond 2011 Programme Director Office for National Statistics, UK.
Arun Srivastava. Small Areas What is a small area? Sub - population Domain The Domain need not necessarily be geographical. Examples Geographical Subpopulations.
Joint UNECE/Eurostat Meeting on Population and Housing Censuses (28-30 October 2009) Accuracy evaluation of Nuts level 2 hypercubes with the adoption of.
Better Information for Regional Government Marie Cruddas, Minda Phillips & Pete Brodie, ONS. Presented by Martin Brand, ONS Methodology Directorate.
Economics and Statistics Administration U.S. CENSUS BUREAU U.S. Department of Commerce Research on Estimating International Migration of the Foreign-Born.
Plans for Access to UK Microdata from 2011 Census Emma White Office for National Statistics 24 May 2012.
Slide 1 Estimating Performance Below the National Level Applying Simulation Methods to TIMSS Fourth Annual IES Research Conference Dan Sherman, Ph.D. American.
MEASURING INCOME AND POVERTY AT A LOCAL LEVEL Sian Rasdale Social Justice Analysis, Scottish Government.
1 Measuring Uncertainty in Population Estimates at Local Authority Level Ruth Fulton, Bex Newell, Dorothee Schneider.
1 Data Linkage for Educational Research Royal Statistical Society March 19th 2007 Andrew Jenkins and Rosalind Levačić Institute of Education, University.
The new multiple-source system for Italian Structural Business Statistics based on administrative and survey data Orietta Luzi, Ugo Guarnera, Paolo Righi.
Evaluating Transportation Impacts of Forecast Demographic Scenarios Using Population Synthesis and Data Simulation Joshua Auld Kouros Mohammadian Taha.
Simon Power Managing Consultant John Rae Director Understanding Communities Through PayCheck
JOINT UN-ECE/EUROSTAT MEETING ON POPULATION AND HOUSING CENSUSES GENEVA, MAY 2009 DETERMINING USER NEEDS FOR THE 2011 UK CENSUS IAN WHITE, Office.
1 Understanding and Measuring Uncertainty Associated with the Mid-Year Population Estimates Joanne Clements Ruth Fulton Alison Whitworth.
Assessing the accuracy of different models for combining aggregate level administrative data Dilek Yildiz Supervisors: Peter W. F. Smith, Peter G.M. van.
United Nations Workshop on Revision 3 of Principles and Recommendations for Population and Housing Censuses and Evaluation of Census Data, Amman 19 – 23.
Poverty, ethnicity and social networks - how are they related? Dharmi Kapadia, Nissa Finney & Simon Peters The University of Manchester The State of Social.
IAOS Shanghai – Reshaping Official Statistics Some Initiatives on Combining Data to Support Small Area Statistics and Analytical Requirements at.
Sinclair Sutherland Labour supply: Finding and using statistics.
Small area estimation combining information from several sources Jae-Kwang Kim, Iowa State University Seo-Young Kim, Statistical Research Institute July.
Statistical Research Update Becky Tinsley Louise Morris.
The evolution of the England and Wales census in a European context Garnett Compton, ONS RSS Conference, 9 September 2015.
Weighting and imputation PHC 6716 July 13, 2011 Chris McCarty.
Looking for statistical twins
Census Planning and Management for next Nigerian Census
2011 Census Results.
Detecting and understanding interviewer effects on survey data using a cross-classified mixed-effects location scale model Ian Brunton-Smith, University.
Evaluating the potential for moving away from a traditional census Becky Tinsley Office for National Statistics (ONS), UK.
Business in the Community Race Equality Campaign
Mesfin S. Mulatu, Ph.D., M.P.H. The MayaTech Corporation
Deriving a reliable measure of household income – DWP
A Comparison of Two Nonprobability Samples with Probability Samples
Addressing the Health SDGs The Challenge of Disaggregation
Worklessness Data on Neighbourhood Statistics
MUSIC EDUCATION RESEARCH SUMMARY
Antidepressant Use Among Working Age Canadians:
Some challenges for small area estimation
Scope for Decentralization of Land Administration in Africa: Evidence from Local Administrative Data in Mozambique Raul Pitoro Michigan State University,
Patterns and trends in adult obesity
Regression composite estimation for the Finnish LFS from a practical perspective Riku Salonen.
Adjusting Census Figures
CPD Programme for Policing Data Specialists Fundamentals
Tabulations and Statistics
Small area estimation of violent crime victim rates in the Netherlands
The European Statistical Training Programme (ESTP)
Estimation of Employment for Cities, Towns and Rural Districts
Albania 2021 Population and Housing Census - Plans
The European Statistical Training Programme (ESTP)
Chapter: 9: Propensity scores
ANALYSIS OF POSSIBILITY TO USE TAX AUTHORITY DATA IN STS. RESULTS
Salah Merad Methodology Division, ONS
JOINT UN-ECE/EUROSTAT MEETING ON POPULATION AND HOUSING CENSUSES
New Techniques and Technologies for Statistics 2017  Estimation of Response Propensities and Indicators of Representative Response Using Population-Level.
Marie Reijo, Population and Social Statistics
Andrew Jenkins and Rosalind Levačić
The role of metadata in census data dissemination
Small area estimation with calibration methods
European Conference on Quality in Official Statistics
Pete Benton , Beyond 2011 Programme Director
Population Statistics without a Census or Register
Chapter 13: Item nonresponse
SMALL AREA ESTIMATION FOR CITY STATISTICS
Presentation transcript:

Small Area Estimation Programme Dr Alison Whitworth

Outline History of SAE at ONS SAE and Census Transformation Small Area Statistics Problem Successful ONS experiences on Small Area Estimation SAE and Census Transformation Best use of administrative sources Lessons learned and challenges

Small Area Statistics Problem Requirement for: Comprehensive, timely and reliable information Data for detailed geographical areas or subcategories But Financial and operational constraints Pressure to reduce survey sample sizes and respondent burden Solution! Borrow information from other related datasets from similar areas, previous occasions

Borrowing strength Small area estimation methods: Modelling procedure Use statistical models that relate the survey data with auxiliary information (borrow strength) Auxiliary information: administrative data or census data available for all areas/domains Modelling procedure Variable of interest -> survey data (Dependent variable) Independent variables -> auxiliary data (covariates) Model based estimates: predicted values used for obtaining area/domain estimates

Successful ONS experiences Small Area Estimation at ONS begun as a research project in 1990s Small Area Estimation Project (SAEP), 1998 EURAREA, 2001-2004 Applications: Unemployment (GB)– annual LAD and PCA estimates updated quarterly Mean Income (E&W) – ward/MSOA estimates 98/99, 01/02, 04/05 and 07/08 Census – 2001 One Number Census, 2011

Mean household weekly income SAEP method (Heady et al. 2003) Estimates for Middle layer Super Output Area (MSOA) Survey data: Family Resources Survey (clustered household sample survey) Auxiliary data: Social benefit claimants, Income tax, data, council tax banding, Census variables …. Linear regression model for log income Uses unit level response and area level covariates

2011-2012 Official model-based MSOA estimates (England and Wales) 95% Confidence Intervals for net income, equivalised, after housing costs

Limitations of the ONS current approach (SAEP) for household income Only allows for estimation of mean household income in each MSOA User requirement for mean and percentiles estimates at lower level geographies

Research in progress in the SAE Unit Examples : Estimation of Household Income Distribution for 2001 and 2011 Using the Empirical Best Predictor (EBP) Method Population Estimates by Local Authority and Ethnic Group Using Generalised Structure Preserving Estimators (GSPREE)

Example 1: Household Income using EBP approach Empirical Best Predictor (Molina and Rao, 2010) Estimation of mean and percentiles of income and poverty measures under one framework Unit level model (household & area level covariates) Involves prediction of income for each household in the population Requires access to household level census data in addition to survey data (non-census years?)

2001 EBP estimates North West and South East regions of England Income Distribution across four different MSOAs

Coefficients of variation for income for five percentiles of the population 2001 EBP estimates North West and South East regions of England

Conclusions EBP approach seems promising MSOA estimates of income distribution and poverty measures under one methodology Still more questions! How to apply EBP approach in non-census years

Example 2: Population by Ethnic Group using GSPREE Motivation: National Statistician’s recommendation: make the best use of all available data in the production of population statistics. Governments ambition: Censuses after 2021 be conducted using other sources of data… Census Transformation programme Research the potential use of administrative data and surveys to produce population, household and characteristic information currently provided in a Census. Generalise Structure Preserving Estimation (GSPREE) 14

Census Table Population by Local Authority and Ethnic Group England (March 2011) Local Authority White Mixed Asian Chinese Black Other Total Fareham 107959 1359 1200 467 357 239 111581 Southampton 203528 5678 16443 3449 5067 2717 236882 Portsmouth 181182 5467 9863 2611 3777 2156 205056 …. … Tower Hamlets 114819 10360 96392 8109 18629 5787 254096 Slough 64053 4758 54900 797 12115 3582 140205 …..

Data for Ethnic Group 2011 Census estimates (Mar 2011) Proxy: Detailed cross tabulation but outdated School Census (Jan 2014) Proxy: Detailed cross tabulation but age 5-15 only Annual Population Survey (2014) Total population by ethnic group Mid Year Population Estimates (2014) Total population by local authority

Solution… Combine administrative and census data with survey data to borrow strength and produce reliable estimate for each cell (domain) using GSPREE (Zhang and Chambers, 2004 and Luna-Hernandez, A, 2014).

Data for Ethnic Group MYE 2014 Census 2011   MYE 2014 White Mixed Asian Chinese Black Other Total Fareham 107959 1359 1200 467 357 239 …….. Southampton 203528 5678 16443 3449 5067 2717 Portsmouth 181182 5467 9863 2611 3777 2156 …..... …. … Tower Hamlets 114819 10360 96392 8109 18629 5787 Slough 64053 4758 54900 797 12115 3582 ……… ….. APS July 2012 - June 2014 (weighted estimates) National total ………. ……….. School Census Dec 2014   White Mixed Asian Chinese Black Other Fareham … Southampton Portsmouth …. Tower Hamlets Slough …..

Distribution of LA estimates by ethnic group, 2014 (England)

Proportion of White – GSPREE & Census

RMSE. LA by ethnic group, 2014 Fixed Effects GSPREE estimator (England) Overall, GSPREE is successful in providing reliable estimates for most LAs. However, non-negligible RMSEs (and CVs) are observed in some areas

Conclusions GSPREE shows good performance Work in progress Small bias/RMSE in most LAs Work in progress Validation study (1991/2001 Census) GSPREE: 2001 Census x 2011 data (APS, MYE, ESC) Validation: 2011 Census Still more questions… Modelling strategy for more detailed categories Discuss alternatives to generate the synthetic population (bootstrap) Consider different attributes

Academic support Collaborative projects Structure Preserving Estimation (SPREE), Expert advisory group Small area estimation Funded research/ Bids e.g. NCRM Conferences NCRM Bath (July) SAE Maastrict (Aug)

Lessons learned Good small area estimates depend on: Challenges: adequacy of the modelling procedures + covariates with good prediction power model validation Challenges: ability to master the complexities of the required statistical theory availability of relevant administrative/auxiliary data capacity to overcome barriers for the acceptance of model based estimates as official statistics outputs

References Molina, I. and Rao, J. N. K. (2010). Small area estimation of poverty indicators. The Canadian Journal of Statistics 38, 369-385 Purcell, N. J. and Kish, L. (1980). Postcensal Estimates for Local Areas (or Domains). International Statistical Review, 48, 3-18. Zhang, L.C. and Chambers, R. (2004). Small area estimates for cross-classifications. Journal of the Royal Statistical Society, B, 66, 479–496. Luna-Hernandez, A. (2014). On Small Area Estimation for Compositions Using Structure Preserving Models. Unpublished PhD upgrade document, Department of Social Statistics and Demography, University of Southampton.