Help for ANOVA from Package replicateBE [🇷 for BE/BA]

posted by Weidson – Brazil, 2021-12-30 22:16 (818 d 18:05 ago) – Posting: # 22720
Views: 2,925

Dear friends,


By studying the replicateBE package documentation https://github.com/Helmut01/replicateBE#readme I was able to verify that the method.A and method.B functions (package replicateBE) are derived from the functions lm and lme (packages stats and nlme).

If possible I would like to reproduce the same ANOVA tables that are displayed in the output of the methods.A and method.B using only the lm and lme functions as indicated in the instructions. Unfortunately I am not reaching my goal of obtain the same results as expected:no:. The hipotetical data set that I working is from replicated 2x4 study and the ANOVA results to be different (GL, p-values, etc...). What am I doing wrong? Why can't I generate the same ANOVA Table? The my results to be bellow:

Somebody can help me?

I Would like publish at forum my data set that generates this results however I didn't see option for upload files excel (just figures). Is it possible publish excel files? If yes, How?

R Code 1.0
method.A with IPC_Drug.Cmax Data Set


method.A(path.in = path, path.out = path, file = paste0(fn, ".", "Cmax"),  ext = "csv", na = "NA", sep = sep, dec = dec, ola = TRUE, print = TRUE, verbose = TRUE, ask = TRUE, plot.bxp = TRUE)

ANOVA Outputs for R Code 1.0:

Data set IPC_Drug.Cmax: Method A by lm()
────────────────────────────────────────
Type III Analysis of Variance Table

Response: log(PK)
                           Df    Sum Sq       Mean Sq    F value      Pr(>F)
sequence                   1      1.3253      1.32531    0.27224    0.604233
period                     3      3.5824      1.19413    2.19909    0.090673
treatment                  1      2.7923      2.79228    5.14222    0.024818
sequence:subject           48   233.6721      4.86817    8.96514    < 2e-16
Residuals                 146    79.2796      0.54301                 

treatment T – R:
  Estimate Std. Error    t value   Pr(>|t|)
-0.2380370  0.1049710 -2.2676500  0.0248184
146 Degrees of Freedom


R Code 1.1
Obtain ANOVA by lm() function with DS_Cmax Data Set


library(stats)
Fit_A <- lm(log(Cmax) ~ Sequence + Subject%in%Sequence + Period + Treatment, data=DS_Cmax)
anova(Fit_A)


ANOVA Outputs for R Code 1.1:

Analysis of Variance Table

Response: log(Cmax)
                        Df       Sum Sq           Mean Sq           F value       Pr(>F)
Sequence                 1       1.32531015       1.325310155       0.83233     0.36273
Period                   1       0.31093557       0.310935573       0.19528     0.65905
Treatment                1      3.41095080        3.410950796       2.14218     0.14492
Sequence:Subject         2      6.70138392        3.350691961       2.10433     0.12470
Residuals                194   308.90305612       1.592283794


R Code 2.0
method.B with IPC_Drug.Cmax Data Set


method.B(path.in = path, path.out = path, file = paste0(fn, ".", "Cmax"), ext = "csv", na = "NA", sep = sep, dec = dec, ola = TRUE,  print = TRUE, verbose = TRUE, ask = TRUE, plot.bxp = TRUE)

ANOVA Outps for utR Code 2.0:

Data set IPC_Drug.Cmax: Method B (option = 2) by lme()
──────────────────────────────────────────────────────
Response: log(PK)
                numDF       denDF       F-value              p-value
(Intercept)      1           146      763.258972853        <.0001
sequence         1            48      0.272239993          0.6042
period           3           146      2.199092956          0.0907
treatment        1           146      5.142222444          0.0248

treatment T – R:
    Value Std.Error   t-value   p-value
-0.238040  0.104970 -2.267600  0.024818
146 Degrees of Freedom (equivalent to SAS’ DDFM=CONTAIN)


R Code 2.1
Obtain ANOVA by lme() function with DS_Cmax Data Set


library(nlme)
Fit_B_opt2 <- lme (log(Cmax)~ Sequence +  Period + Treatment, random = ~1|Subject, data=DS_Cmax)
anova(Fit_B_opt2)


ANOVA Outputs for R Code 2.1:

               numDF       denDF       F-value            p-value
(Intercept)     1           148       763.258972846       <.0001
Sequence        1            48         0.272239993       0.6042
Period          1           148         0.561664145       0.4548
Treatment       1           148         6.161433208       0.0142

Complete thread:

UA Flag
Activity
 Admin contact
22,957 posts in 4,819 threads, 1,638 registered users;
86 visitors (0 registered, 86 guests [including 13 identified bots]).
Forum time: 16:22 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