Quantcast
Channel: Statalist
Viewing all articles
Browse latest Browse all 65555

non-inferiority hypothesis testing

$
0
0
I am interested in using a ttest to test: Ho: Mean(VSI)>Mean(VO)+8 OR EQUIVALENTLY Mean(VSI)-Mean(VO)-8>0 and Ha: Mean(VSI)<=Mean(VO)+8 OR Mean(VSI)-Mean(VO)-8<=0

I understand that the lower right-hand section of the ttest is the non-inferiority pvalue, but are are able to specify the bounds/non-inferiority limits in Stata? Specifically, I would like to specify the difference of 8 in this test. Thank you.

ttest avg_compress_depth_family, by(mode_of_teaching_logistic) level(90)

Two-sample t test with equal variances
------------------------------------------------------------------------------
Group | Obs Mean Std. Err. Std. Dev. [90% Conf. Interval]
---------+--------------------------------------------------------------------
0 | 285 45.77681 .6545839 11.05065 44.69659 46.85703
1 | 237 40.16034 .7639371 11.76066 38.89882 41.42185
---------+--------------------------------------------------------------------
combined | 522 43.2268 .5124008 11.70698 42.38248 44.07113
---------+--------------------------------------------------------------------
diff | 5.616469 1.000274 3.968229 7.26471
------------------------------------------------------------------------------
diff = mean(0) - mean(1) t = 5.6149
Ho: diff = 0 degrees of freedom = 520

Ha: diff < 0 Ha: diff != 0 Ha: diff > 0
Pr(T < t) = 1.0000 Pr(|T| > |t|) = 0.0000 Pr(T > t) = 0.0000



Viewing all articles
Browse latest Browse all 65555

Trending Articles



<script src="https://jsc.adskeeper.com/r/s/rssing.com.1596347.js" async> </script>