search

timriffe / DemoTools

Tools for the evaluation, adjustment, and standardization of demographic data
https://timriffe.github.io/DemoTools
Other
61 stars 31 forks source link

Progress on Graduation vignette. #88

Closed jmaburto closed 4 years ago

jmaburto commented 4 years ago

Progress on Graduation vignette. Should we use some specific example for the 'rescaleAgegroups'function? if so which ones?

What do you mean by graduation as a light smoother>

Graduation as a lifetable step. Should we include this in the graduation vignette or in the lifetable vignette?

timriffe commented 4 years ago

  1. rescaleAgeGroups() I'm, not sure. How about you get DemoData death or pop counts from an HMD country and in single ages from Jorge. Then rescale them to match HMD counts in 5-year age groups? Just an idea.

  2. The graduation functions work strictly with data in 5-year age groups. So data will be grouped to 5-year ages on the fly if necessary. So the combo of grouping and then graduating ends up being a kind of constrained smooth. It's light compared to the agesmth family because there aren't transfers between 5-year age groups.

  3. I don't know, it's kind of niche. Maybe that's it's own fun vignette for this specific topic for when pkgdown is operational. The idea is that graduation followed by single age lifetable will produce an implied nAx, which can be backed out. The alternative is to do something sophisticated to get nAx well behaved at the level of abridged lifetables. Graduation is just a cheap way to avoid those steps, because once you're in single ages it's not so bad to use midpoints as ax.