jimjohn wrote:
> hi guys, just a quick question. if i want a confidence itnerval for a
> lognormally distributed variable, can i still use the standard normal
> distribution table to calculate my critical values. or are the critical
> values different for lognormals than for normals? thanks!
>
Compute the CI with the log transformed data, and then back-transform
the limits. Remember that you will have then a CI for the geometric
mean, not the arithmetic mean.
Marta GG
--
For miscellaneous statistical stuff, visit:
http://gjyp.nl/marta/=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD
Locked
