Skip to content

New estimates (data and do files updates; no code updates)#210

Closed
pbronka wants to merge 1 commit intosimpaths:developfrom
pbronka:bugfix/merge-reverted-changes
Closed

New estimates (data and do files updates; no code updates)#210
pbronka wants to merge 1 commit intosimpaths:developfrom
pbronka:bugfix/merge-reverted-changes

Conversation

@pbronka
Copy link
Copy Markdown
Contributor

@pbronka pbronka commented Jul 4, 2025

These changes were originally made directly on the develop branch but without corresponding code changes. We decided to revert these data changes from the develop branch until the required changes are made to the code (by reverting @dariaple's original commit), however doing so prevents opening a pull request from @dariaple's original branch as these changes are already in the history.

This applies the following changes made by @dariaple:

Summary of updates:
•Estimation sample and initial populations now cover the years 2011–2023.
•Added year-specific dummies (y2020, y2021) to all processes to capture the impact of the pandemic.
•Ethnicity variables:
-Included the four-category variable dot in all processes.
-Added an extended ethnicity variable dot01 to initial populations with six categories.
•Replaced dhe with its physical (dhe_pcs) and mental (dhe_mcs) health components wherever health was included.
•Replaced disability dummy dlltsd with dlltsd01, which captures both self-reported disability and receipt of disability benefits.
•Weights: All processes now use dimxwt (Individual Cross-Sectional Weight).
•Automatised the production of Excel files with estimates.
•There are process-specific updates detailed in the Excel files with estimates.
•Added an Excel file summarising the structure of covariates used in the estimates.

Note
This PR should only be merged after the required code changes to accommodate the data changes above have been made.

This reverts commit 71e4e05.
@pbronka
Copy link
Copy Markdown
Contributor Author

pbronka commented Jul 8, 2025

TODO:

  1. Ethnicity:
  • Keep 6-category ethnicity variable in the initial populations and to evolve in the simulation. Move mixed white to white category, so mixed becomes mixed non-white.
  • Aggregate up to 4 categories on the fly when required by the regression models
  • Children inherit ethnicity from mothers
  1. Dhe_pcs and dhe_mcs:
  • Merge regression models predicting dhe_pcs and dhe_mcs into the code (from Daniel?)
  1. Dlltsd01:
  • Drop the dlltsd

@pbronka
Copy link
Copy Markdown
Contributor Author

pbronka commented Aug 8, 2025

This PR looks redundant since #214 has been added

@pbronka pbronka closed this Aug 8, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants