stata command for upper tail confidence interval

[gracey inno]

Construct an upper 95% confidence interval for the mean of full score IQ for lead exposure group 1, and a lower 95% confidence interval for the mean of full score IQ for lead exposure group 2 (medium exposure). This is the question but i'm having a hard time finding the stata command for upper and lower 95% confidence interval. Any help will be much appreciated

[Clyde Schechter]

The -ci- command will enable you to calculate confidence intervals.

To get a 95% one-sided confidence interval, use the -level(90)- option. An upper confidence interval would be from the lower limit of that result to infinity. A lower confidence interval would be from the upper limit of that result to negative infinity.

[gracey inno]

thank you, but if i type level 90 for both, how will stata differentiate between the upper and lower level

[Clyde Schechter]

Stata won't. You have to do that. For the 95% upper confidence interval you just ignore the upper limit and report the interval from the lower limit to infinity. For the 95% lower confidence interval, you do the opposite.

[gracey inno]

thank you