Bioequivalence and Bioavailability Forum • EMA dataset 1

EMA dataset 1 - programmers trouble [RSABE / ABEL]

posted by d_labes – Berlin, Germany, 2013-04-09 11:01 (4467 d 10:29 ago) – Posting: # 10379
Views: 6,101

Hi ElMaestro,

❝ Here's some code bits:

❝

❝ A=read.table("EMAset1b.csv", header=T)

❝ ....

❝ lnU=lnPE+s*CritT*sqrt(0.5*(1/nBABA+1/nABAB)) ##can't get it right.

❝ lnL=lnPE-s*CritT*sqrt(0.5*(1/nBABA+1/nABAB)) ##can't get it right.

Try it with:

lnU=lnPE+s*CritT*sqrt(0.25*(1/nBABA+1/nABAB))

lnL=lnPE-s*CritT*sqrt(0.25*(1/nBABA+1/nABAB))

or even easier fit a model with intercept and get the confidence interval via call of confint(). No need to fiddle with n in sequences:

all <- lm(log(Data) ~ as.factor(formulation)+ as.factor(period) +as.factor(subject), data=A)

lnPE <- coef(all)[2]


CI   <- confint(all,level=0.9)[2,]

BTW: I have omitted sequence since it doesn't contribute to the mse to have such a term. Moreover I have all effects in lower letters (caps lock not arrested :-D

).

My SAS gives (without back transformation):

                              Standard

  Parameter    Estimate          Error   t Value   Pr > |t|     90% Confidence Limits



  test-ref   0.14547372     0.04650869      3.13     0.0020     0.06864574   0.22230170

The latter R code gives:

> print(lnPE)

as.factor(tmt)T 

      0.1454737 

> print(CI)

       5 %       95 % 

0.06864569 0.22230165

—
Regards,

Detlew

Complete thread:

EMA dataset 1 trouble ElMaestro 2013-04-08 20:23
- complete (RR) only Helmut 2013-04-09 02:37
- EMA dataset 1 - programmers troubled_labes 2013-04-09 09:01
  - Still not exactly sure ElMaestro 2013-04-09 10:12
    - Still not exactly sure - why d_labes 2013-04-09 15:57
      - My problem(s) ElMaestro 2013-04-09 16:39
      - More confusion ElMaestro 2013-04-09 23:29
        
        Confusion help d_labes 2013-04-10 10:51