TODO list - Githubissues

aaowens / PSID.jl

Quickly assemble data from the Panel Study of Income Dynamics (PSID)

MIT License

25 stars 9 forks source link

TODO list #22

Open aaowens opened 4 years ago

aaowens commented 4 years ago

Test more, increasing code coverage. Might need to generate some fake data.
Change the Dict key type to String so I can update JSON3 (edit: Done)

aaowens commented 4 years ago

It also would be nice to be able to pull information directly from the individual file. There is some useful information in there, like education (missing in some years in the family files), and the individual sampling weights.

eirikbrandsaas commented 4 years ago

Here are some more suggestions, features that would be nice to include:

[ ] Include support for supplementary files
- [ ] Wealth Files
[ ] Speed up the reading of the .txt files
[ ] Allow for only using a subset for years. For example, limit the sample to the period (1999-) became biennial. (This also would speed up the reading of the text files)

aaowens commented 4 years ago

Ya, I think using the supplementary files would require adding an extra field to the input.json file to indicate which files (family, individual, wealth, other?) to look in, and then some code to use that.