sudy
☆    

India,
2017-08-01 15:29
(2431 d 00:57 ago)

Posting: # 17641
Views: 5,028
 

 Inter subject CV calculation [General Sta­tis­tics]

Hi all,
Can anyone suggest any logic on below results?

[image]


Why Inter subject CV is not calculated for Cmax?
Is this formula is correct to calculate between subject variability on log transformed data for two way cross over study?
If i calculate manually in excel, I am getting the value around 28% for cmax. But SAS gives error.
One more point i have noted that, the MSS is very low by SAS, but manually this value is very high.
Why R-square is very low for Cmax?


Edit: Category changed; see also this post #1. [Helmut]
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2017-08-01 16:33
(2430 d 23:53 ago)

@ sudy
Posting: # 17642
Views: 4,361
 

 √x, where x<0

Hi sudy,

next time you upload a PNG please 640px (longest side) or better only the numbers. I had a hard time to read the results. :cool:

❝ Why Inter subject CV is not calculated for Cmax?


Unless you want to deal with imaginary numbers, the square root of a negative value is not defined. Since ℯ(MSs – MSe)/2 = 0.994155, √–0.005845 don’t work.

❝ Is this formula is correct to calculate between subject variability on log transformed data for two way cross over study?


Yes. See there.

❝ If i calculate manually in excel, I am getting the value around 28% for cmax. But SAS gives error.


Please check what you have done in Excel. Cannot work:

A1 0.06596695
B1 0.07769188
C1 =(A1-B1)/2
D1 =EXP(C1)
E1 =D1-1
F1 =100*SQRT(E1)

What do you get? 28% is CVintra – in Excel-lingo 100*SQRT((EXP(B1)-1)).

❝ One more point i have noted that, the MSS is very low by SAS, but manually this value is very high.


No idea. I don’t have a crystal ball which could tell me what you are doing “manually”.

❝ Why R-square is very low for Cmax?


Higher variability than the ones of AUCs ⇒ lower R2. As expected.

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
UA Flag
Activity
 Admin contact
22,957 posts in 4,819 threads, 1,638 registered users;
62 visitors (0 registered, 62 guests [including 10 identified bots]).
Forum time: 15:26 CET (Europe/Vienna)

Nothing shows a lack of mathematical education more
than an overly precise calculation.    Carl Friedrich Gauß

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