This package uses DataDeps.jl to download the US distributional national accounts (DINA) dataset from Gabriel Zucman's website. The data are downloaded in bulk (>1GB) and stored in .julia/datadeps/dina/
.
The dataset is described in Piketty, T, E. Saez and G. Zucman: Distributional National Accounts: Methods and Estimates for the United States, Quarterly Journal of Economics, 2018, 133 (2): 553-609.
The variables are described in the codebook at Gabriel Zucman's website.
Load the DINA micro-data by year
using DINA, DataFrames
tbl = get_dina(1980)
df80 = DataFrame(tbl)
or load a quantile-year panel (takes about 1-2 minutes for 1962--2019).
using DINA
inc = :peinc
wgt = :dweght
dbt_var = [:ownermort, :rentalmort, :nonmort, :hwdeb]
var = [[:fiinc, :fninc, :ptinc, inc, :poinc, :ownerhome, :rentalhome]; dbt_var]
byvar = inc # compute deciles of `:peinc`
df = dina_quantile_panel(var, byvar, 10)
This is a picture you can produce using df
. Have a look at the documentation for the full code.