24.01.2013 Views

handout - Daina Chiba

handout - Daina Chiba

handout - Daina Chiba

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

2.2.2 Putting Stars<br />

For the sake of convenience, I think it is good to highlight statistically significant coefficients<br />

in some ways. I have no objection to putting stars, but keep in mind that there are some<br />

people who think putting stars is a sin. Some of those people who hate stars, however, think<br />

that it is OK to highlight statistically significant coefficients by writing them in thick font<br />

(I honestly don’t know why). If, unfortunately, your reviewer or your TA expresses such a<br />

belief, you’d better listen to him than trying to point out the stupidity of that person.<br />

FYI, AJPS prohibits us from placing multiple stars to differentiate multiple confidence<br />

levels.<br />

2.3 Exporting Stata Results to R<br />

This will become essential when you need to estimate some fancy statistical models that are<br />

easier to implement in Stata, but then want to use R to make graphs.<br />

Run the following chunk of Stata code:<br />

Stata code<br />

insheet using "output/smw06162010dataS.txt", clear<br />

global covariates "smldem lrgdem smldep allies lncaprat dircont lndstab majpower"<br />

replace smldep = smldep*1000<br />

logit mzmid1 $covariates<br />

mat beta = e(b)<br />

mat vcov = e(V)<br />

// Export Beta and V-COV matrix<br />

preserve<br />

svmat beta, names(matcol)<br />

outsheet beta* in 1 using "output/sr_beta.txt", replace nolabel<br />

restore<br />

preserve<br />

svmat vcov, names(matcol)<br />

outsheet vcov* in 1/9 using "output/sr_vcov.txt", replace nolabel<br />

restore<br />

Two text files will be created (replaced) within the output folder.<br />

9

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!