Login or Register
Log in with
Forums
FAQ
Search in titles only
Search in General only
Advanced Search
Search
Home
Forums
Forums for Discussing Stata
General
You are not logged in. You can browse but not post. Login or
Register
by clicking 'Login or Register' at the top-right of this page. For more information on Statalist, see the
FAQ
.
Announcement
Collapse
No announcement yet.
X
Collapse
Posts
Latest Activity
Search
Page
of
1
Filter
Time
All Time
Today
Last Week
Last Month
Show
All
Discussions only
Photos only
Videos only
Links only
Polls only
Filtered by:
Clear All
new posts
River Huang
Join Date:
Mar 2016
Posts:
1887
#1
Why not significant?
13 Mar 2019, 19:39
Dear All, I was asked why the estimate below (shaded) with t-value -2.04 is not statistically significant? Any comments? Thanks.
Ho-Chuan (River) Huang
Stata 17.0, MP(4)
Rich Goldstein
Join Date:
Mar 2014
Posts:
4267
#2
13 Mar 2019, 19:43
you only have 3 clusters - do not use "cluster(varname)" with that small a number
Comment
Post
Cancel
River Huang
Join Date:
Mar 2016
Posts:
1887
#3
13 Mar 2019, 19:51
Thanks, Rich. Agreed, but is this related to the problem?
Ho-Chuan (River) Huang
Stata 17.0, MP(4)
Comment
Post
Cancel
Joseph Coveney
Join Date:
Apr 2014
Posts:
4137
#4
13 Mar 2019, 20:08
Originally posted by
River Huang
View Post
I was asked why the estimate below (shaded) with t-value -2.04 is not statistically significant? Any comments?
Three clusters gives you two degrees of freedom.
. display in smcl as text 2 * t(2, -abs(-2.04))
.1781673
.
Comment
Post
Cancel
River Huang
Join Date:
Mar 2016
Posts:
1887
#5
13 Mar 2019, 20:17
Dear Joseph, Thanks so much. This helps.
Ho-Chuan (River) Huang
Stata 17.0, MP(4)
Comment
Post
Cancel
Previous
Next
Working...
Yes
No
OK
OK
Cancel
X
Comment