CV2 replenishing population

[ld-archer]

Cross validation 2 is where we use all the data in waves 1-4 for estimating transition models, and simulate from wave 5-8 using these transition models.

The comparison population for the CV2 'handover' plots is built from ELSA_long, which therefore includes the replenishing waves in the survey (3, 4...). We need to make sure we generate our plots using a comparative population. There are a few things to take into account:

• Replenishing population in ELSA
  • Age limit? i.e. only plot data for those aged 55+. This would nullify the replenishing pop.s in ELSA
  • Include replenishing in the simulation. Would need to generate a replenishing from ELSA_stock_CV1
• Match population by id?
  • Guarantee that we are comparing the exact same population.

First try the age limit and matched population, as these two are easy to implement together.

• [ ] Age limit in plots
  • How young? People will be simulated for 4 waves (8 years), so minimum age would have to be 59? Or 58?
• [ ] Match population by ID before plotting

Note: The CV2 output summary file does not allow us to get outputs by age, so will need to use detailed output instead.

[ld-archer]

Would be helpful to have a before and after.

These are the handover plots BEFORE making the above changes: