Title: RE: Confidence interval
At 4:58 PM -0400 9/2/03, Hildreth, Bruce wrote:
Hi Bruce, see embedded comments
-----Original Message-----
From: Bruce Jackson [mailto:[EMAIL PROTECTED]
Sent: Tuesday, September 02, 2003 3:37 PM
Cc: Bruce Hildreth
Subject: Confidence interval

During this month's working group telecon, the topic of confidence intervals for aero data came up. I think I misspoke then about using a single value for "confidence," and wanted to clarify my thoughts.

The current draft AIAA standard for simulation data exchange


recommends carrying, for any data table so desired, a set of points showing the 95% confidence bound (with the table values themselves being the nominal value to use if not performing a Monte Carlo or stress test).
I[Hildreth, Bruce]  thought 3 sigma was 95%.  However, the confidence interval info should support symmetric and assymetric distrubutions of any confidence,  We should add to the DTD;
1)global distrubution specification definition  ( ex. normal, binomial, uniform... etc)
2)global sigma (accuracy) specification (ex. 1, 3, 5 whatever... etc)
and the related two DTD for a specific function that overrides the global definition
I do not believe that the spec should require a specific definition, but needs to accomodate whatever the data is.  So I think we agree
Note:  for symmetric distributions only one value needs to be set,  Only for assymetric distributions do you need the + and the -.

I am posting this reply to [EMAIL PROTECTED] so others may benefit from your wisdom, Bruce H. :).

I don't see an exceptional utility to a global distribution specification; this is only a small amount of data to append to each table. I agree we need both a distribution specification and an accuracy specification (the popular "six sigma" will be the favorite of management types!) but I think it is best left to reside with individual tables.

Bruce Jackson mailto:[EMAIL PROTECTED] Dynamics and Control Branch
18C West Taylor Street MS 132            Airborne Systems Competency
NASA Langley Research Center                Building 1192C, Room 149
Hampton, VA 23681-0001            scud://N 37 05'31.7" W 76 22'55.1"
http://dcb.larc.nasa.gov/DCBStaff/ebj/ebj.html     "There is no try"

Reply via email to