## Parallel bears meeting at random in infinity [R for BE/BA]

Dear all, especially dear bears,

since I'm

If I got it right the code used is (f.i. AUC log-transformed):

I am wondering where this code comes from, what this code does an why it works anyhow .

IMHO this model, one (fixed) effect for the treatment and one (random) effect for the subjects must be over-specified. We have only one value for a distinct subject treated with Test or Reference and thus we are not able to separate this uniquely into 2 effects, one part for treatment and one attributed to the subject.

Nevertheless the lme() call produces a result.

Try it out in bear.

If I follow the strange and crude EMA suggestion and use all effects as fixed

with the bear built-in dataset for parallel groups I get the anticipated result:

BTW: I would go for a parallel groups study with exactly

Is anybody out there who knows a generalization of the Welch test to more than 2 groups? Any hint would be very appreciated!

Or could we use pairwise Welch t-tests for that?

since I'm

**very**interested in using R for evaluation of BE studies I had a closer look at the bear code for parallel group studies (**as part of validation ).***code inspection*If I got it right the code used is (f.i. AUC log-transformed):

`lme(lnAUC0t ~ drug, random=~1|subj, data=TotalData, method="REML" )`

I am wondering where this code comes from, what this code does an why it works anyhow .

IMHO this model, one (fixed) effect for the treatment and one (random) effect for the subjects must be over-specified. We have only one value for a distinct subject treated with Test or Reference and thus we are not able to separate this uniquely into 2 effects, one part for treatment and one attributed to the subject.

Nevertheless the lme() call produces a result.

Try it out in bear.

If I follow the strange and crude EMA suggestion and use all effects as fixed

`lmModel <- lm(lnAUC0t ~ drug + subj, data=TotalData)`

with the bear built-in dataset for parallel groups I get the anticipated result:

```
>summary(lmModel)
```

Call:

lm(formula = lnAUC0t ~ drug + subj, data = TotalData)

Residuals:

ALL 20 residuals are 0: no residual degrees of freedom! (over-specified!)

Coefficients: (1 not defined because of singularities)

Estimate Std. Error t value Pr(>|t|)

(Intercept) 7.3199 NA NA NA

drug2 -0.0622 NA NA NA

subj2 -0.5366 NA NA NA

[...]

>anova(lmModel)

Analysis of Variance Table

Response: lnAUC0t

Df Sum Sq Mean Sq F value Pr(>F)

drug 1 0.00127 0.001270

subj 18 1.17959 0.065533

Residuals 0 0.00000

BTW: I would go for a parallel groups study with exactly

**2**groups with the 'simple' t-test (Welch variant as described by Helmut long ago) .Is anybody out there who knows a generalization of the Welch test to more than 2 groups? Any hint would be very appreciated!

Or could we use pairwise Welch t-tests for that?

—

Regards,

Detlew

Regards,

Detlew

### Complete thread:

- Parallel bears meeting at random in infinity - d_labes, 2010-04-22 11:43 [R for BE/BA]
- Parallel bears meeting at random in infinity - ElMaestro, 2010-04-22 12:53
- Parallel groups in bear - CIs - d_labes, 2010-04-22 14:00
- Parallel groups in bear - CIs - ElMaestro, 2010-04-22 21:47
- Parallel groups in bear - CIs - d_labes, 2010-04-23 09:09

- Parallel groups in bear - CIs - yjlee168, 2010-04-25 23:29

- Parallel groups in bear - CIs - ElMaestro, 2010-04-22 21:47

- Parallel groups in bear - CIs - d_labes, 2010-04-22 14:00
- Parallel bears meeting at random in infinity - yjlee168, 2010-04-22 23:09
- Modelling Parallel bears - d_labes, 2010-04-23 09:12
- Modelling Parallel bears - yjlee168, 2010-04-23 21:14
- Validating vs. WinNonlin... - Helmut, 2010-04-24 00:28
- Validating vs. WinNonlin... - yjlee168, 2010-04-24 19:36
- Validating vs. WinNonlin... - yjlee168, 2010-04-26 00:09
- Validating vs. WinNonlin... - Helmut, 2010-04-26 01:29
- WNL in replicate BE - yjlee168, 2010-04-26 08:59
- WNL in replicate BE - Helmut, 2010-04-26 16:15

- WNL in replicate BE - yjlee168, 2010-04-26 08:59

- Validating vs. WinNonlin... - Helmut, 2010-04-26 01:29

- Modelling Parallel bears - yjlee168, 2010-04-25 19:34
- Modelling Parallel bears - ElMaestro, 2010-04-25 20:40
- Dataset - Helmut, 2010-04-25 22:38
- Dataset - yjlee168, 2010-04-25 22:44
- Dataset - Helmut, 2010-04-26 01:13
- Dataset - yjlee168, 2010-04-26 08:16
- NCA → Statistical analysis for parallel study - Helmut, 2010-04-26 13:12
- NCA → Statistical analysis for parallel study - yjlee168, 2010-04-26 18:43

- NCA → Statistical analysis for parallel study - Helmut, 2010-04-26 13:12

- Dataset - yjlee168, 2010-04-26 08:16

- Dataset - Helmut, 2010-04-26 01:13
- dilemma - yjlee168, 2010-04-26 08:41
- Equal variances - d_labes, 2010-04-26 09:04
- Equal variances - yjlee168, 2010-04-26 09:22
- GLM = Equal variances - d_labes, 2010-04-26 13:29
- GLM = Equal variances - Helmut, 2010-04-26 14:45
- I'm a believer - d_labes, 2010-04-26 15:58
- I'm a believer - Helmut, 2010-04-26 16:31

- I'm a believer - d_labes, 2010-04-26 15:58

- GLM = Equal variances - Helmut, 2010-04-26 14:45

- GLM = Equal variances - d_labes, 2010-04-26 13:29
- Equal variances - Helmut, 2010-04-26 12:55
- gls() for unequal variances? - d_labes, 2010-04-26 16:36
- gls() for unequal variances? - Helmut, 2010-04-26 17:00
- Sims - Helmut, 2010-04-27 01:36
- Sandwich - Simsalabim - d_labes, 2010-04-28 10:58
- Sandwich - Simsalabim - Helmut, 2010-04-28 14:19
- parametrization of R function rlnorm - martin, 2010-05-02 18:22
- Mean of log-normal - d_labes, 2010-05-03 16:22
- parametrization of R function rlnorm - ElMaestro, 2013-07-26 21:42
- Martin‽ - Helmut, 2013-07-28 02:01

- Sandwich - Simsalabim - d_labes, 2010-04-28 10:58

- gls() for unequal variances? - d_labes, 2010-04-26 16:36

- Equal variances - yjlee168, 2010-04-26 09:22

- Dataset - yjlee168, 2010-04-25 22:44

- Validating vs. WinNonlin... - Helmut, 2010-04-24 00:28

- Modelling Parallel bears - yjlee168, 2010-04-23 21:14

- Modelling Parallel bears - d_labes, 2010-04-23 09:12

- Parallel bears meeting at random in infinity - ElMaestro, 2010-04-22 12:53