Bioequivalence and Bioavailability Forum • bear for 2x2x2 study with negative variance components

bear for 2x2x2 study with negative variance components [Software]

posted by yjlee168 – Kaohsiung, Taiwan, 2014-02-05 20:12 (4094 d 02:46 ago) – Posting: # 12349
Views: 28,069
(edited on 2014-02-06 08:02)

Dear all,

Using lme() to a 2x2x2 BE study, we can code with R (taking AUC_0-t as example of the dataset in this thread) something like


modlnAUC0t<-lme(log(AUC0t) ~ drug + seq + prd,

               random=~1|subj/seq,

               data=blablabla, method="REML")

cat("\n")

print(summary(modlnAUC0t))

cat("\n")

cat("Type I Tests of Fixed Effects\n")

print(anova(modlnAUC0t)[2:4,])

cat("\n")

cat("Type III Tests of Fixed Effects\n")

print(anova(modlnAUC0t, type="marginal")[2:4,])

cat("\n\n")

...

Then outputs will be


Linear mixed-effects model fit by REML

 Data: blablabla

       AIC      BIC   logLik

  2.163918 10.41029 5.918041



Random effects:

 Formula: ~1 | subj

         (Intercept)

StdDev: 1.106768e-06   <- (1.11*10^-6)² ≈ 0 = V(subject(seq))



 Formula: ~1 | seq %in% subj

         (Intercept)  Residual

StdDev: 1.100023e-06 0.1562085 <- (0.156)² = 0.0244 = MSE



Fixed effects: log(AUC0t) ~ drug + seq + prd 

                Value  Std.Error DF   t-value p-value

(Intercept)  9.516955 0.05904125 12 161.19163  0.0000

drug2        0.062297 0.05904125 12   1.05514  0.3121

seq2        -0.036450 0.05904125 12  -0.61736  0.5485

prd2        -0.048908 0.05904125 12  -0.82837  0.4236

...

Standardized Within-Group Residuals:

       Min         Q1        Med         Q3        Max 

-1.8446164 -0.7801051  0.1796820  0.6539527  1.6438227 



Number of Observations: 28

Number of Groups: 

         subj seq %in% subj 

           14            14 

            numDF denDF   F-value p-value

(Intercept)     1    12 103679.28  <.0001

drug            1    12      1.11  0.3121

seq             1    12      0.38  0.5485

prd             1    12      0.69  0.4236



Type I Tests of Fixed Effects

     numDF denDF   F-value p-value

drug     1    12 1.1133109  0.3121

seq      1    12 0.3811317  0.5485

prd      1    12 0.6862044  0.4236



Type III Tests of Fixed Effects

     numDF denDF   F-value p-value

drug     1    12 1.1133109  0.3121

seq      1    12 0.3811317  0.5485

prd      1    12 0.6862044  0.4236

Since we will still stick on lm() with 2x2x2 study, so bear will show the outputs as


...

Intra_subj. CV = 100*sqrt(exp(MSResidual)-1) = 18.038 %

Inter_subj. CV = 100*sqrt(exp((MSSubject(seq)-MSResidual)/2)-1)

               = 0 % (with a negative variance component)

*** the above CV_intra is estimated from lm() which may be different

    from than that obtained from lme().



    MSResidual = 0.03201731 

MSSubject(seq) = 0.01678485

Therefore, the CV_intra from lme() will not be calculated in this case. But luckily for us, Detlew has showed us that V(subject(seq)) is zero (or should be close to zero) and the 90%CI was the same as what we got from lm() using SAS PROC MIXED. Sorry for this lengthy post.

—
All the best,
-- Yung-jin Lee
bear v2.9.2:- created by Hsin-ya Lee & Yung-jin Lee
Kaohsiung, Taiwan https://www.pkpd168.com/bear
Download link (updated) -> here

Complete thread:

How to calculate intersubject variability in PHX WinNonlin zan 2014-01-29 23:58 [Software]
- Negative variance component Helmut 2014-01-30 01:16
  - Negative variance component zan 2014-01-30 18:11
  - Negative variance component zan 2014-01-31 00:16
    - Negative variance component ElMaestro 2014-01-31 08:20
    - Negative variance component yjlee168 2014-01-31 10:26
      - Example data set Helmut 2014-02-01 16:03
        
        Example data set yjlee168 2014-02-01 17:40
        
        PHX build 6.3.0.395 / 6.4.0.511 Helmut 2014-02-02 02:04
        
        PHX build 6.3.0.395 / 6.4.0.511 yjlee168 2014-02-02 07:54
    - Negative variance component ElMaestro 2014-02-01 16:31
      - Negative variance component yjlee168 2014-02-01 17:47
        
        Just thinking loud ElMaestro 2014-02-01 19:02
        
        All models are wrong… Helmut 2014-02-02 02:31
        
        another book for linear model yjlee168 2014-02-02 08:07
      - References Helmut 2014-02-02 02:19
        
        References ElMaestro 2014-02-02 09:56
    - Negative variance component – Chow/Liu d_labes 2014-02-03 09:02
      - Negative variance component – Chow/Liu ElMaestro 2014-02-03 10:22
        
        Variance components – Proc mixed d_labes 2014-02-03 11:58
        
        Variance components – Proc mixed ElMaestro 2014-02-03 12:58
        
        Variance components – Proc mixed 90% CIs d_labes 2014-02-03 13:16
        Variance components – Proc mixed Helmut 2014-02-03 14:14
        
        FDA code for non-replicate crossover? d_labes 2014-02-03 15:54
        
        Proc GLM rulez Helmut 2014-02-03 16:16
        
        Variance components – Proc mixed yjlee168 2014-02-03 20:43
        
        NOBOUND Helmut 2014-02-03 22:08
      - lm() or lme() for 2x2x2 study design? yjlee168 2014-02-03 20:22
        
        lm() or lme() for 2x2x2 study design? ElMaestro 2014-02-03 22:11
        
        lm() or lme() for 2x2x2 study design? yjlee168 2014-02-04 13:09
        
        bear for 2x2x2 study with negative variance componentsyjlee168 2014-02-05 19:12