Changing the level of significance in Correlation

[lal mohan kumar]

Dear all,
I would like to get an excel output related to correlation analysis. I ran the following commands

estpost correlate var1 var2 var3  ,matrix listwise
est store c1
esttab * using corrmatr.csv,unstack not noobs compress

Here the default level of significance ,stars are given as follows, "* p<0.05,** p<0.01&*** p<0.001". I would like to change it as * p<.10,** p<.05 &*** p<.01.
How to do this?
Can anyone help me here

[Wouter Wakker]

help esttab##stars

[lal mohan kumar]

Dear Wouter Wakker
I coudnt get it cleared. For instance, I tried

 pwcorr varl var2, star(.05) replace

The above code, reports stars for showing statistical significance at .05. What I want is in one table, how we can generate all-star sign for 0.1, 0.5, and 0.01. Similar to the correlation table published in the articles. Sometimes, in one table, some variables may be significant at 1%, some at 5% and some at 10%.
I also tried

 pwcorr var1 var2, print(.10)star(*** 0.10 ** 0.05 ***.01)
option star() incorrectly specified
r(198)

Any help in this regard will be helpful

[Wouter Wakker]

esttab * using corrmatr.csv, unstack not noobs compress star(* 0.10 ** 0.05 *** 0.01)

In #2 I alluded to the fact that this is documented in the help file of esttab, which you can view by typing help esttab in Stata's command window.

[lal mohan kumar]

Thanks Wouter Wakker, but still the problem is not solved. The excel file displays 3 stars and there are no single star and 2 stars. What I want is one star for significance at .10 level and 2 stars for significance at .05 level and 3 stars for significance at .01 level. That is not the case with code I tried which I have attached

sysuse auto.dta
estpost correlate price mpg rep78 trunk  ,matrix listwise
est store c1
esttab * using corrmatr.csv, unstack not noobs compress star(* 0.10 ** 0.05 *** 0.01)

[Wouter Wakker]

Thank you for the reproducible example. However, in this example there are no correlation coefficients with significance at the 5 or 10 percent level, which is why there are no cases with one or two stars. I don't see the problem.