Progress Update on Demographic Accounts Project for June 2022 Delivery
The update covers the progress of the Demographic Accounts project, including proof-of-concept milestones, annual local authority level estimations, model approaches, and data preprocessing. The project aims to deliver demographic accounts by age, sex, and local authority, incorporating data from various sources to enhance accuracy and forecasting capabilities.
Download Presentation
Please find below an Image/Link to download the presentation.
The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author. Download presentation by click this link. If you encounter any issues during the download, it is possible that the publisher has removed the file from their server.
E N D
Presentation Transcript
Demographic accounts project update Louisa Blackwell1, Duncan Elliott1 and John Bryant2 1Office for National Statistics 2Bayesian Demography Limited 17 May 2022 Official Sensitive (if required)
Overview Update on progress for June 2022 delivery Demographic Accounts proof-of-concept (LA annual) Design Results Data considerations Plans for June 2022 publication Discussion
Proof-of-concept milestones June 2022: Annual 2011-2022 demographic accounts by age, sex and local authority September 2022: monthly July 2011-June 2022 demographic accounts by age and sex (England & Wales) December 2022: monthly July 2011-December 2022 demographic accounts by age, sex and local authority
Update Progress towards June 2022 delivery
Annual local authority level estimation Cohort-oriented state-space framework Parallelisation of estimation Simplified approach (implemented in account package) Calculate demographic rates (smoothed) prior to modelling Birth and death registrations are known Internal, cross-border and international migration combined prior to modelling and split out post modelling with iterative proportional fitting Modelling each LA independently Technical description initial draft gives details
Model approach: June 2022 milestone Use data available for initial proof of concept, forecasting beyond 2021 Simplified assumptions about uncertainty for Statistical Population Datasets Census 2021 can be added to the model but is not essential for meeting the proof-of-concept (ie framework can incorporate/drop different sources)
Pre-processing and checking inputs Raw demographic rates by age, sex, year and local authority are too noisy to be considered superpopulation rates Smoothing using Generalised Additive Models Input from demographic experts for refinement of models
Computational breakthrough empirical testing National level model demest vs account (run-time down from time ~30 hours to ~20 minutes) Running model for all 331 Local Authorities is now feasible and initial tests demonstrate ability to use alternative population stock data as inputs alternative assumptions about data models Simulation study in progress This is still in proof-of-concept stage, further development of models is required
Data considerations Real-time dashboard: timely feeds on data covering population, migration, births and data Informing trends and requirements for projections in the model Summarising and quality assuring large amounts of input and output data Need regular monthly inputs processing and quality issues will be important, eg lags Prioritising inputs needing improvements
Publishing outputs for June 2022 Time series 2011-2022 for selected LAs using SPD V2, with and without Census 2021 in the model. Time series 2011-2022 for selected LAs using SPD V3, with and without Census 2021 in the model. Comparison of 2022 from (i) and (ii) above, (selected LAs with different population dynamics, for visualisation purposes)
Comments and discussion
Project team and collaborators John Bryant, Bayesian Demography Limited University of Southampton: Peter WF Smith, Jakub Bijak and Jason Hilton ONS: Louisa Blackwell, Claire Bloss, Duncan Elliott, Gillian Fleetwood, Valentina Gribanova, Claire Jewell, Beck Keane, Jennifer Langer, Salah Merad, Aidan Metcalfe, Luke Morris, Rob North, Greg Payne, Sonya Ridden, Arunn Sathasivam, Curtis Sinclair Integrated Statistical Design Working Group: demographers across ONS
![Project Initiation Document for [Insert.Project.name] [Insert.Project.number]](/thumb/226757/project-initiation-document-for-insert-project-name-insert-project-number.jpg)
