GNU addiction [General Sta­tis­tics]

posted by Helmut Homepage – Vienna, Austria, 2011-10-02 05:20 (5018 d 10:35 ago) – Posting: # 7406
Views: 15,982

Dear ElMaestro!

❝ A great paper comparing certain numeric precision aspects here.


THX for reminding about this paper. Had to give Gnumeric (1.10.16) another try. Wow! Have a look at its functions. tinv() does not round dfs!
df     tinv(0.05,df)
5      2.5705818356364616
5.123  2.5521289049159468
6      2.4469118511451238


There are two implementations of the inverse β-distribution: betainv() for Excel/OO Calc compatibility and r.qbeta() calling R’s algo. However, results are identical:
betainv(0.001,2,4)=r.qbeta(0.001,2,4)=0.010101787883737754

Now to your example:
[image]

Looking at the formulas behind it’s clear: Welcome to Welch’s test!
[image]

        Welch Two Sample t-test

data:  log(TEST) and log(REF)
t = 0.0596, df = 3.876, p-value = 0.9555
alternative hypothesis: true difference in means is not equal to 0
90 percent confidence interval:
 -0.1196929  0.1265074
sample estimates:
mean of x mean of y
 3.568160  3.564752



[image] Rulez!

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
23,424 posts in 4,927 threads, 1,687 registered users;
73 visitors (0 registered, 73 guests [including 8 identified bots]).
Forum time: 15:56 CEST (Europe/Vienna)

Complex, statistically improbable things are by their nature
more difficult to explain than
simple, statistically probable things.    Richard Dawkins

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