Cross-over 3x6 WNL x SAS [Software]
I'm trying to get the same results with SAS, for the WinNonlin output, for a cross-over design 3x6 (Chow & Liu, page 321, PURICH's data), without success.
Someone tell me how to get the same results with SAS code?
Following the data, the SAS program and output, and the WinNonlin output.
WinNonlin output:
SAS Program:
SAS Output:
Mainly the period effect is very different!
Thanks and sorry about my English!!
Edit: Category changed. Please don't use tabulators in you post (see here on how to use BBCodes) - use the Preview before posting. [Helmut]
Someone tell me how to get the same results with SAS code?
Following the data, the SAS program and output, and the WinNonlin output.
Subj Per SEQ Droga asc Carry lnasc
1 1 RTS 1 5.68 1 1.736951
2 1 RTS 1 3.6 1 1.280934
3 1 SRT 2 3.55 1 1.266948
4 1 SRT 2 7.31 1 1.989243
5 1 TSR 3 6.59 1 1.885553
6 1 TSR 3 9.68 1 2.270062
7 1 STR 2 4.63 1 1.532557
8 1 STR 2 8.75 1 2.169054
9 1 TRS 3 7.25 1 1.981001
10 1 TRS 3 5 1 1.609438
11 1 RST 1 4.63 1 1.532557
12 1 RST 1 3.87 1 1.353255
1 2 RTS 3 4.21 1 1.437463
2 2 RTS 3 5.01 1 1.611436
3 2 SRT 1 5.07 2 1.623341
4 2 SRT 1 7.42 2 2.004179
5 2 TSR 2 7.72 3 2.043814
6 2 TSR 2 8.91 3 2.187174
7 2 STR 3 7.23 2 1.978239
8 2 STR 3 7.59 2 2.026832
9 2 TRS 1 7.88 3 2.064328
10 2 TRS 1 7.84 3 2.059239
11 2 RST 2 6.77 1 1.912501
12 2 RST 2 7.62 1 2.030776
1 3 RTS 2 6.83 3 1.921325
2 3 RTS 2 5.78 3 1.754404
3 3 SRT 3 4.49 1 1.501853
4 3 SRT 3 7.86 1 2.061787
5 3 TSR 1 7.26 2 1.98238
6 3 TSR 1 9.04 2 2.201659
7 3 STR 1 5.06 3 1.621366
8 3 STR 1 4.82 3 1.572774
9 3 TRS 2 9.02 1 2.199444
10 3 TRS 2 7.79 1 2.052841
11 3 RST 3 5.72 2 1.743969
12 3 RST 3 6.74 2 1.90806
WinNonlin output:
Partial Tests of Model Effects
Hypothesis Numer_DF Denom_DF F_stat P_value
------------------------------------------------------------------------
int 1 6 335.277 0.0000
seq 5 6.27 1.66382 0.2711
droga 2 18 3.69067 0.0454
Per 2 18 3.28399 0.0608
Carry 2 18 1.21341 0.3204
SAS Program:
PROC GLM DATA=purich;
CLASS seq subj per droga carry;
MODEL lnauc = droga per seq subj(seq) carry /SS3;
RANDOM subj(seq)/test;
RUN;
QUIT;
SAS Output:
The GLM Procedure
Tests of Hypotheses for Mixed Model Analysis of Variance
Dependent Variable: lnauc
Source DF Type III SS Mean Square F Value Pr > F
Droga 2 0.230110 0.115055 3.44 0.0543
Per 2 0.085202 0.042601 1.27 0.3037
Subj(SEQ) 6 0.702524 0.117087 3.50 0.0179
Carrytest 2 0.081128 0.040564 1.21 0.3204
Error: MS(Error) 18 0.601883 0.033438
Source DF Type III SS Mean Square F Value Pr > F
SEQ 5 0.874041 0.174808 1.58 0.2896
Error 6.3004 0.695535 0.110395
Error: 0.92*MS(Subj(SEQ)) + 0.08*MS(Error)
Mainly the period effect is very different!
Thanks and sorry about my English!!
Edit: Category changed. Please don't use tabulators in you post (see here on how to use BBCodes) - use the Preview before posting. [Helmut]
Complete thread:
- Cross-over 3x6 WNL x SASMaura 2010-09-02 17:09
- WNL partial F vs. SAS type III F d_labes 2010-09-03 08:58
- WNL partial F vs. SAS type III F Helmut 2010-09-03 11:13
- WNL almost the same as SAS d_labes 2010-09-03 12:10
- WNL almost the same as SAS Maura 2010-09-03 14:07
- WNL almost the same as SAS d_labes 2010-09-03 12:10
- WNL partial F vs. SAS type III F Helmut 2010-09-03 11:13
- WNL partial F vs. SAS type III F d_labes 2010-09-03 08:58