Diagnostics: R [Study As­sess­ment]

posted by Helmut Homepage – Vienna, Austria, 2016-05-23 01:22  – Posting: # 16349
Views: 20,055

Hi zizou,

» DF, p-values, CIs not reported in lmer for the reasons stated there.

Yep, we know. :-D

» library lmerTest can be used,

THX; I forgot!

» I think there is a little bug in df visualization - rounded on 3 decimal places, visible on 5 (actually I have not the latest version of R).

Same in R 3.2.5 and lmerTest 2.0-30.

» summary(muddle)$coefficients["(Intercept)","df"]
»  # [1] 9.195607


Or easier:
summary(muddle)$coefficients[, "df"]
(Intercept)   log(Dose)
   9.195607    5.896110


summary(muddle, digits=8) doesn’t help. Or as Carl Witthoft wrote: IMHO the proper use of summary is indicated by its name: a way to get a quick-look at your data.

» For calculation of 90% CIs: (below is only example for Intercept lower and upper limit)
» alpha=0.05
» summary(muddle)$coefficients["(Intercept)","Estimate"] - qt(1-alpha,summary(muddle)$coefficients["(Intercept)","df"]) * summary(muddle)$coefficients["(Intercept)","Std. Error"]

Perfect! Little bit shorter:
summary(muddle)$coefficients[1, 1] +c(-1,+1)*qt(1-0.05, summary(muddle)$coefficients[1, 3]) * summary(muddle)$coefficients[1, 2]
[1] 1.496758 2.386013
summary(muddle)$coefficients[2, 1] +c(-1,+1)*qt(1-0.05, summary(muddle)$coefficients[2, 3]) * summary(muddle)$coefficients[2, 2]
[1] 0.6695768 0.8539044


So what do we have? A fair agreement across software – except SAS

Software               Estimate       90% CI       width
NCSS               B0 1.9414    1.4849    2.3978    0.9129
                   B1 0.7617    0.6659    0.8576    0.1917
Phoenix/WinNonlin  B0 1.9413858 1.4967592 2.3860125 0.8893
                   B1 0.7617406 0.6695783 0.8539029 0.1843
R lmerTest         B0 1.9413858 1.4967584 2.3860132 0.8893
                   B1 0.7617406 0.6695768 0.8539044 0.1843
SAS (code?)        B0 1.94      1.54      2.35      0.8100
                   B1 0.7615    0.679     0.844     0.1650

Since I trust most in Phoenix and R; given the width of the CI: Is NCSS too conservative and SAS liberal? Unfortunately we don’t have Smith’s code. Would be great if one of our SASians could jump in.

Cheers,
Helmut Schütz
[image]

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

Complete thread:

Activity
 Admin contact
20,463 posts in 4,297 threads, 1,415 registered users;
online 5 (2 registered, 3 guests [including 3 identified bots]).
Forum time (Europe/Vienna): 01:01 CEST

Half of the modern drugs could well be thrown out of the window,
except that the birds might eat them.    Martin H. Fischer

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