One-sided confidence intervals PreviousNext
Mplus Discussion > Structural Equation Modeling >
Message/Author
 Marius Wrulich posted on Tuesday, September 27, 2011 - 10:33 am
Dear Dr. Muthén,

I would like to ask if there is any possibility to obtain one-sided 95% confidence intervals for analyses using the Mplus 5.2 software. Specifically, we hope to obtain one-sided CIs for coefficients from straightforward linear regression and ordinal logistic regression, as we are testing one-tailed hypotheses.

One way we thought of might be to take the range of the two-sided 95% confidence interval from the Mplus output, then determine the lower (or, depending our hypothesis, the upper) bound of a two-sided 90% CI, and then add (or substract) the range of the two-sided 95% to (or from) this bound.

However, if I understand correctly, Mplus only gives two-sided 95% and 99% CIs for regression parameters, not 90% CIs.

Do you perhaps have any other suggestions on how to obtain one-sided CIs for regression parameters?

Thank you very much for your help.
 Linda K. Muthen posted on Tuesday, September 27, 2011 - 2:04 pm
Mplus does not provide one-sided confidence intervals. You would need to compute those yourself.
Back to top
Add Your Message Here
Post:
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Password:
Options: Enable HTML code in message
Automatically activate URLs in message
Action: