Closed alnahedh closed 8 years ago
Hi there,
What I would do is to actually bootstrap the R-squared values as well, here is how:
parallel setclusters 4
sysuse auto
parallel bs, rep(1000) exp(_b e(r2)): reg price foreign rep78 mpg
Which returns something like this:
--------------------------------------------------------------------------------
Parallel Computing with Stata (by GVY)
Clusters : 4
pll_id : qw4s1kfp49
Running at : /home/george
Randtype : datetime
Waiting for the clusters to finish...
cluster 0004 has exited without error...
cluster 0001 has exited without error...
cluster 0003 has exited without error...
cluster 0002 has exited without error...
--------------------------------------------------------------------------------
Enter -parallel printlog #- to checkout logfiles.
--------------------------------------------------------------------------------
parallel bootstrapping Number of obs = 69
Replications = 1000
command: regress price foreign rep78 mpg
[_eq2]_bs_1: e(r2)
------------------------------------------------------------------------------
| Observed Bootstrap Normal-based
| Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
_eq1 |
foreign | 1023.208 650.8838 1.57 0.116 -252.5008 2298.917
rep78 | 432.8004 300.4363 1.44 0.150 -156.0439 1021.645
mpg | -292.4342 62.81709 -4.66 0.000 -415.5535 -169.315
_cons | 10586.48 1610.768 6.57 0.000 7429.438 13743.53
-------------+----------------------------------------------------------------
_eq2 |
_bs_1 | .2667079 .0898997 2.97 0.003 .0905077 .4429081
------------------------------------------------------------------------------
So, if you use outreg2, you get
. outreg2 using parallelbs, replace
Which looks something like this
(1) (2)
VARIABLES _eq1 _eq2
foreign 1,023
(650.9)
rep78 432.8
(300.4)
mpg -292.4***
(62.82)
_bs_1 0.267***
(0.0899)
Constant 10,586***
(1,611)
Observations 69 69
Standard errors in parentheses
*** p<0.01, ** p<0.05, * p<0.1
Hope it helps!
Hello, is there a way to display or extract out (outreg or esttab) the R-Squared when running a "parallel bs" routine? I'm running a basic regression of the form
parallel bs, reps(1000): reg y x1 x2
All I'm getting on the results screen is the number of obs and replications..Help or future implementation of this is appreciated..
Thanks, Saad