[R] confidence intervals

Ebert,Timothy Aaron tebert @end|ng |rom u||@edu
Mon Aug 29 03:50:13 CEST 2022

I have a general dislike of "analysis emergencies." I would like to see a data emergency wherein someone must cram 3 years of data collection into 18 months so that they have time to work out the correct analysis. I am sure others would suggest working out how analyze the data before starting the experiment.

Our business office gives this advice to faculty members: An emergency on your part is not an emergency on our part. 

How about starting by answering the questions posted by the people you are hoping will help. Focus on David's middle paragraph. However, if you can re-code everything to work, then it would seem that you already know the answer and it might be simpler/faster to write the correct code.

You might spend some time looking for a scientific paper that uses that equation for the confidence interval and thereby get some context to explain why the equation is correct.


-----Original Message-----
From: R-help <r-help-bounces using r-project.org> On Behalf Of Bogdan Tanasa
Sent: Sunday, August 28, 2022 8:55 PM
To: David Winsemius <dwinsemius using comcast.net>
Cc: r-help <r-help using r-project.org>
Subject: Re: [R] confidence intervals

[External Email]

Hi David,

Thank you for your comments, and feed-back message. I am very happy to learn from the experience of the people on R mailing list, and without any doubt, I am very thankful to you and to everyone for sharing their knowledge. I do apologize for any confusion that I have created unwillingly with my previous email.

About my previous email related to the confidence intervals: indeed I have posted the question with a detailed description on stackoverflow, and the link is listed below.

I have to admit that I have been in rush willing to have the suggestions of R-help members by Monday (if that would have been possible), as I have to make a decision at the beginning of this week on whether I need to re-code the shell script in R. I have a deadline on Wed. The script itself is less important per se, I have included it just to point our the origin of my question.

I do certainly respect the principles of online R-help community, and I would very much appreciate if I could have your advice on the following :
shall a "R code related emergency" arise, would it be acceptable to post the question on stackoverflow with the corresponding data tables and detailed code, and to refer the posting on R-help mailing list ?

If it is acceptable at least for a single email, and if you do not mind, I could mention the link to stackoverflow, inviting our members to read it, shall they be comfortable with this topic.


Thanks a lot, have  a good week !

~ Bogdan


On Sat, Aug 27, 2022, 6:52 PM David Winsemius <dwinsemius using comcast.net>

> You cross-posted this to StackOverflow and did not say so.  ... and 
> you posted in HTML Bad dog squared. I cast one of the close votes on 
> SO, but here I can only say ... READ the Posting Guide.
> You also give no citation other than someone's Github files with 
> minimal comments in that material. You should indicate whether this 
> code has any solid support. Why do you think this code is something to depend upon?
> After all, you been posting questions on R-help for several months.
> Don't you think you should make a good faith effort to understand the 
> principles underlying this resource?
> --
> David.
> On 8/26/22 17:55, Bogdan Tanasa wrote:
> > Dear all,
> >
> > Although I know that it is not a statistics mailing list, given my 
> > work
> on
> > ICeChIP
> >
> >
> https://nam10.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgith
> ub.com%2Fshah-rohan%2Ficechip%2Fblob%2Fmaster%2FScripts%2FcomputeHMDan
> dError&data=05%7C01%7Ctebert%40ufl.edu%7C0ba5d535471b46c05ec508da8
> 9592c20%7C0d4da0f84a314d76ace60a62331e1b84%7C0%7C0%7C63797331334389431
> 3%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6
> Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=9HOH72K0Lj30w7tE%2BMN
> oXakPbO1sasaujUTsnHFwDZo%3D&reserved=0
> >
> > I would appreciate to have the answer to a question :
> >
> > given two variables a and b (a and b can have 1000 paired-values) 
> > and a calibration number "cal",
> >
> > why the 95 confidence interval has been calculated as such for each 
> > value
> > a(i) and b(i) :
> >
> > 100 / cal * sqrt (( a/ (b^2) + (a^2) / (b ^3)) * 1.96
> >
> > Thank you,
> >
> > Bogdan
> >
> >       [[alternative HTML version deleted]]
> >
> > ______________________________________________
> > R-help using r-project.org mailing list -- To UNSUBSCRIBE and more, see
> > https://nam10.safelinks.protection.outlook.com/?url=https%3A%2F%2Fst
> > at.ethz.ch%2Fmailman%2Flistinfo%2Fr-help&data=05%7C01%7Ctebert%4
> > 0ufl.edu%7C0ba5d535471b46c05ec508da89592c20%7C0d4da0f84a314d76ace60a
> > 62331e1b84%7C0%7C0%7C637973313344050547%7CUnknown%7CTWFpbGZsb3d8eyJW
> > IjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C300
> > 0%7C%7C%7C&sdata=KfmZ6PvxUVrbCMzVZcz6S2H2sqfWQmw2WIZHERwKDEk%3D&
> > amp;reserved=0
> > PLEASE do read the posting guide
> https://nam10.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.r
> -project.org%2Fposting-guide.html&data=05%7C01%7Ctebert%40ufl.edu%
> 7C0ba5d535471b46c05ec508da89592c20%7C0d4da0f84a314d76ace60a62331e1b84%
> 7C0%7C0%7C637973313344050547%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwM
> DAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&
> sdata=127HvzZh7q9Xa7LaESHIEXIy3CdmUyQ3cxAVeOA%2Bukg%3D&reserved=0
> > and provide commented, minimal, self-contained, reproducible code.

        [[alternative HTML version deleted]]

R-help using r-project.org mailing list -- To UNSUBSCRIBE and more, see
PLEASE do read the posting guide https://nam10.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.r-project.org%2Fposting-guide.html&data=05%7C01%7Ctebert%40ufl.edu%7C0ba5d535471b46c05ec508da89592c20%7C0d4da0f84a314d76ace60a62331e1b84%7C0%7C0%7C637973313344050547%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=127HvzZh7q9Xa7LaESHIEXIy3CdmUyQ3cxAVeOA%2Bukg%3D&reserved=0
and provide commented, minimal, self-contained, reproducible code.

More information about the R-help mailing list