search

uchicago-computation-workshop / luis_bettencourt

5 stars 0 forks source link

Why is the US data representative? #18

Open weijia1995 opened 7 years ago

weijia1995 commented 7 years ago

It's very interesting to see the conclusion that 'income group dependent with richer and poorer households appearing spatially more segregated than middle-income groups'. Yet I noticed the data only includes the US metropolitan areas. It may indeed take local economic heterogeneity into account but merely in the US region.

I wonder why the US data is representative because, to my understanding, the income distribution, spatial variation and factors influencing migration and economics in other countries may be different from that of the US's, though the approach is 'general and applicable to any heterogeneous population where individual traits are sorted differentially into groups'. For example, would European data give the same result?

lmbett commented 7 years ago

Good question, it's an empirical question to be investigate. The data for South Africa and Brazil in the other paper - for example - allows you to test that idea, but we have not yet done so...