OT: Spreadsheet addiction [Software]

posted by Helmut Homepage – Vienna, Austria, 2019-07-19 16:29 (1936 d 23:46 ago) – Posting: # 20403
Views: 21,128

Hi Shuanghe,

❝ ❝ Yep. I wrote a package for this stuff.

❝ Well, you're always several steps ahead of me.


Nope. I started on July 11th.

❝ My output is for 4 methods only (Z-distribution, homo- and hetero-scedastic and pragmatic) …


Mine only for homo- and heteroscedastic variances (the latter is the package’s default). I think that relying on large sample approximations doesn’t make sense at all. The CI will always be narrower than with the other approaches but why would one go this way? Approximations of degrees of freedom for unequal variances were published in 1946/47 (Satterthwaite, Welch), the t-distribution in 1908 (Gosset a.k.a. Student). Going back more than two than two centuries (Laplace, Poisson, Gauß)? I don’t get it. BTW, I don’t like the pragmatic method.

❝ … since I'm still struggling with Chow and Shao/Liu's methods.


You need the raw data, right?

❝ […] I didn't see the package in your Github page. Any plan to put it there so we can have a peek? :-D


Paid job – chained by a CDA. If we will publish sumfink, sure. Not my decision.

❝ ❝ If you want to reproduce Excel’s wacky “commercial rounding”:

❝ ❝ ...

❝ I doubt I'll very use it but nice to know that it can be done. ;-)


I would never ever use it. However, if you want to reproduce sumfink which was rounded in Excel you can only experiment till you get the same in R.

❝ ❝ […] navigate up to my OP and hit ctrlF5 to clear the browser’s cache / reload. Is it possible in Excel?


❝ No way. Now you just want to piss off those Excel lovers :lol3:


Yessir! R is a language and environment for statistical computing and graphics.

Dif-tor heh smusma 🖖🏼 Довге життя Україна! [image]
Helmut Schütz
[image]

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes

Complete thread:

UA Flag
Activity
 Admin contact
23,288 posts in 4,890 threads, 1,658 registered users;
31 visitors (1 registered, 30 guests [including 6 identified bots]).
Forum time: 15:16 CET (Europe/Vienna)

‘Normal science’ means research firmly based upon one or more
past scientific achievements, achievements that some particular
scientific community acknowledges for a time as supplying
the foundation for its further practice.    Thomas S. Kuhn

The Bioequivalence and Bioavailability Forum is hosted by
BEBAC Ing. Helmut Schütz
HTML5