neilalex / ad-replication

Paper Replication: The Growth of Low-Skill Service Jobs and the Polarization of the U.S. Labor Market (AER 2013)
https://neilalex.com/paper-replication-the-growth-of-low-skill-services-jobs-and-the-polarization-of-the-us-labor-market/
MIT License
1 stars 3 forks source link

Table 7 #7

Open IoannisPapadakis opened 6 years ago

IoannisPapadakis commented 6 years ago
  1. In this line it seems to me that it should be table7_k'_index' . I do not understand what the following line does : naxSaveOutput "Table6 7 index=index' k=k'" "rtiShare".

  2. It seems that you have skipped the wage analysis. That is, Panel B of table 7.

neilalex commented 6 years ago

1: Yeah good question -- that line saves the coefficients and standard errors from the regression in the previous line to a new row in a dataset called outputs.dta (done via the naxSaveOutput macro. In general, outputs.dta keeps track of these kinds of outputs from all the different tables).

The Table6 7 index=`index' k=`k' part just gives that new line a description (with `index' being the 1, 2, or 3 from this line, and `k' one of the values from this line -- maybe that's obvious).

All these things said, I think the description should start with "Table7 A" not "Table6 7" -- probably a typo.

2: Yes unfortunately I wasn't able to replicate Table 7 Panel B, due to insufficient personal hardware haha. More recently I've been using Stata for cloud / AWS, which I think could resolve that, so perhaps I'll try again

IoannisPapadakis commented 6 years ago

Thanks, @neilalex! The confusing part of the syntax is the last bit "rtiShare". Is this a way to rename the data-file? In naxMerge, for instance, you also have "1 0 1" or "0 0 1" between the last bit (which seems like a name) and the previous bits of the syntax.

neilalex commented 6 years ago

rtiShare is the specific regression variable whose estimated coefficient and standard error we want to save; the regression on line 56 includes a host of variables, though rtiShare is what we want to report on.

So, all in all we have "Table6 7 index=`index' k=`k'", which is the description mentioned above, and "rtiShare", which is the regression variable whose beta and SE we output to outputs.dta.

Does that seem to help? Let me know if I still misunderstood?

Regarding the "1 0 1" syntax in naxMerge, does what I wrote here help at all?

IoannisPapadakis commented 6 years ago

Thanks Neil!