Open vdquadros opened 5 years ago
Year | Group | Code and data | Overall | Residual |
---|---|---|---|---|
1980 | Native men | Card's paper | 0.385 | 0.288 |
Card's code with our dataset | 0.387 | 0.292 | ||
Stata code with our dataset | 0.379 | 0.283 | ||
Native women | Card's paper | 0.317 | 0.269 | |
Card's code with our dataset | 0.316 | 0.272 | ||
Stata code with our dataset | 0.315 | 0.267 | ||
Immigrant men | Card's paper | 0.444 | 0.321 | |
Card's code with our dataset | 0.444 | 0.322 | ||
Stata code with our dataset | 0.435 | 0.327 | ||
Immigrant women | Card's paper | 0.343 | 0.291 | |
Card's code with our dataset | 0.343 | 0.292 | ||
Stata code with our dataset | 0.342 | 0.295 |
If you run regressions on the dataset built from the sas code...what do things look like? (We could imagine building the dataset using sas and then converting to stata...).
Sent from my phone. Apologies for excessive terseness.
On Mar 4, 2019, at 3:13 PM, vdquadros notifications@github.com wrote:
Year Group Code and data Overall Residual 1980 Native men Card's paper 0.385 0.288 Card's code with our dataset 0.387 0.292 Stata code with our dataset 0.379 0.283 Native women Card's paper 0.317 0.269 Card's code with our dataset 0.316 0.272 Stata code with our dataset 0.315 0.267 Immigrant men Card's paper 0.444 0.321 Card's code with our dataset 0.444 0.322 Stata code with our dataset 0.435 0.327 Immigrant women Card's paper 0.343 0.291 Card's code with our dataset 0.343 0.292 Stata code with our dataset 0.342 0.295 — You are receiving this because you were assigned. Reply to this email directly, view it on GitHub, or mute the thread.
I will edit this comment later to explain what this regression is in more detail, but it is basically re-running the regression after I fixed a bug I found while running the files in SAS. As you can see, it pretty much doesn't change Table 6.
------------------------------------------------------------------------------------
1 2 3 4
b/se b/se b/se b/se
------------------------------------------------------------------------------------
Log rel supply imm~e -0.020*** -0.020*** -0.025*** -0.026***
(0.00) (0.00) (0.00) (0.00)
Log msa size 1980 -0.076*** -0.077*** -0.083*** -0.086***
(0.00) (0.00) (0.00) (0.00)
Log msa size 1990 0.077*** 0.078*** 0.086*** 0.089***
(0.00) (0.00) (0.00) (0.00)
College share 1980 0.062*** 0.067*** 0.077*** 0.089***
(0.02) (0.02) (0.02) (0.02)
College share 1990 -0.034** -0.040** -0.039** -0.054***
(0.01) (0.01) (0.01) (0.01)
Wage res native 1980 0.134*** 0.129*** 0.134*** 0.122***
(0.00) (0.00) (0.00) (0.00)
Wage res imm 1980 -0.159*** -0.154*** -0.161*** -0.148***
(0.00) (0.00) (0.00) (0.00)
Mfg share in 1980 -0.067*** -0.057*** -0.094*** -0.072***
(0.01) (0.01) (0.01) (0.01)
Mfg share in 1990 -0.032** -0.046*** -0.005 -0.036**
(0.01) (0.01) (0.01) (0.01)
Lagged dep var -0.011*** -0.026***
(0.00) (0.00)
constant -0.054*** -0.054*** -0.078*** -0.079***
(0.00) (0.00) (0.00) (0.00)
------------------------------------------------------------------------------------
r2 0.144 0.144 0.139 0.138
------------------------------------------------------------------------------------
* p<0.05, ** p<0.01, *** p<0.001
.
.
. estout m3 m4 m7 m8, cells(b(star fmt(3)) se(par fmt(2))) ///
> legend label varlabels(_cons constant) ///
> stats(r2)
------------------------------------------------------------------------------------
5 6 7 8
b/se b/se b/se b/se
------------------------------------------------------------------------------------
Log rel supply imm~e -0.056*** -0.049*** -0.076*** -0.066***
(0.00) (0.00) (0.00) (0.00)
Log msa size 1980 -0.021*** -0.024*** -0.038*** -0.035***
(0.00) (0.00) (0.00) (0.00)
Log msa size 1990 0.012*** 0.020*** 0.034*** 0.034***
(0.00) (0.00) (0.00) (0.00)
College share 1980 -0.211*** -0.216*** -0.163*** -0.180***
(0.02) (0.02) (0.02) (0.02)
College share 1990 0.071*** 0.111*** 0.048*** 0.085***
(0.01) (0.01) (0.01) (0.01)
Wage res native 1980 0.208*** 0.312*** 0.213*** 0.291***
(0.01) (0.01) (0.01) (0.01)
Wage res imm 1980 -0.123*** -0.238*** -0.122*** -0.210***
(0.00) (0.00) (0.00) (0.00)
Mfg share in 1980 -0.392*** -0.437*** -0.406*** -0.436***
(0.01) (0.01) (0.01) (0.01)
Mfg share in 1990 0.516*** 0.563*** 0.509*** 0.546***
(0.01) (0.01) (0.01) (0.01)
Lagged dep var 0.179*** 0.135***
(0.00) (0.00)
constant -0.049*** -0.081*** -0.128*** -0.131***
(0.00) (0.00) (0.00) (0.00)
------------------------------------------------------------------------------------
r2 0.350 0.378 0.322 0.361
------------------------------------------------------------------------------------
* p<0.05, ** p<0.01, *** p<0.001
Table 2
Looks exactly the same as Card's.
Table 3
Table 6
Columns (1)-(4) are still a bit strange, especially because the R2 is lower and the sign for the lagged dependent variable is negative (instead of positive as in Card's paper). Columns (5)-(8) are pretty close.