Comparing 99th percentiles

Moderators: statman, Analyst Techy, andris, Fierce, GerineL, Smash

anandak
Posts: 14
Joined: Sat Oct 06, 2007 9:04 am

Comparing 99th percentiles

Postby anandak » Fri Apr 25, 2008 3:53 pm

Hi, I was wondering if anyone can help me with this problem:

How do you determine whether the 99th percentile of a distribution is significantly different (at p < .05) from the 99th percentile of another distribution?

All help deeply appreciated

PAK
statman
Administrator
Posts: 2750
Joined: Tue Jun 12, 2007 12:08 pm
Location: Florida, USA

Postby statman » Fri Apr 25, 2008 8:33 pm

While not specifically at any percentile, the chi sq can test association
See the note below

NOTE: Please read the Posting Guidelines and always tell us your OS, the SPSS version and information about your study and data!

Statman
Statistical Services
anandak
Posts: 14
Joined: Sat Oct 06, 2007 9:04 am

99th percentile

Postby anandak » Mon Apr 28, 2008 12:57 pm

Hi Statman,

Thanks for your reply. Chi-square test is for testing association between two non-parametric variables. Can it be used in my case since the 99th percentile is from a interval variable.

Moreover we are comparing the 99th percentile of distribution/ variable A with the 99th percentile of another distribution/variable B - this is a test of difference between them. Thought we should use a parametric test but do not know which and how.

Any help appreciated.

PAK

Who is online

Users browsing this forum: No registered users and 1 guest

cron