BookmarkSubscribeRSS Feed
☑ This topic is solved. Need further help from the community? Please sign in and ask a new question.
AndersS
Lapis Lazuli | Level 10

Hi! I need to calculate estimates of the confidence intervals for extreme quantiles, say q1 and q99 (or q(0.1) and q(99.9)).

My knowledge of this type of mathematical statistics is not so great.
I know that there are options in Proc Means or Univariate, for something like this.

 

But I need to do these calculations in a Data-step.

If anyone has a tip of some good papers to read and implement, then I would appreciate thsi very much.

 

(I will use this info to try to automatically estimate the parameters LLM and ULIM in the method: 
   "Fast and Accurate Calculation of Descriptive Statistics of Very Large Sets of Data".

The work is paused at present. I need a break. But extended paper will appear during 2024).

 

Many thanks in advance!
/Br AndersS (Anders Sköllermo)

Anders Sköllermo (Skollermo in English)
1 ACCEPTED SOLUTION

Accepted Solutions
Rick_SAS
SAS Super FREQ

The CIPCTLDF option on the PROC UNIVARIATE statement produces distribution-free confidence intervals. The formulas are in the doc.

 

Hahn and Meeker are a little vague about one portion of the formula. See this discussion from someone who was trying to reproduce the formula in R and was unable to match the PROC UNIVARIATE output.

View solution in original post

7 REPLIES 7
PaigeMiller
Diamond | Level 26

Why would you want to use a DATA step if SAS has already done the work and provided confidence intervals for percentiles in PROC UNIVARIATE (and probably elsewhere)?

--
Paige Miller
AndersS
Lapis Lazuli | Level 10

Hi! 1) se my text in ( ).
2) See the answer from Rick.  I want to be able to use another language than SAS.

/Br AndersS

Anders Sköllermo (Skollermo in English)
Rick_SAS
SAS Super FREQ

The CIPCTLDF option on the PROC UNIVARIATE statement produces distribution-free confidence intervals. The formulas are in the doc.

 

Hahn and Meeker are a little vague about one portion of the formula. See this discussion from someone who was trying to reproduce the formula in R and was unable to match the PROC UNIVARIATE output.

AndersS
Lapis Lazuli | Level 10

Hi! The basic discussion was in 2019. Has SAS updated the software and/or the documentation? 

 

VERY useful discussion for me! MANY thanks! /Br AndersS

Anders Sköllermo (Skollermo in English)
Rick_SAS
SAS Super FREQ

> The basic discussion was in 2019. Has SAS updated the software and/or the documentation? 

The software has not changed. I gave the link to the most recent documentation.

AndersS
Lapis Lazuli | Level 10

Hi! Very good. My problem has been solved. /Br AndersS

Anders Sköllermo (Skollermo in English)

SAS Innovate 2025: Call for Content

Are you ready for the spotlight? We're accepting content ideas for SAS Innovate 2025 to be held May 6-9 in Orlando, FL. The call is open until September 25. Read more here about why you should contribute and what is in it for you!

Submit your idea!

How to Concatenate Values

Learn how use the CAT functions in SAS to join values from multiple variables into a single value.

Find more tutorials on the SAS Users YouTube channel.

Click image to register for webinarClick image to register for webinar

Classroom Training Available!

Select SAS Training centers are offering in-person courses. View upcoming courses for:

View all other training opportunities.

Discussion stats
  • 7 replies
  • 1491 views
  • 4 likes
  • 4 in conversation