Log in
Register|
jag009 ★★★ NJ, 2013-04-08 22:18 (4419 d 08:31 ago) Posting: # 10373 Views: 5,176 |
|
|
Hi, Does anyone have trouble trying out the 4way full replicate SAS code from FDA's progesterone guidance? I tried it with the EMEA 4-way datasets but I am stuck at the merging routine for "test1 test2 ref1 ref2" and the ilat and dlat calculation. Determine Iij andDij data scavbe; merge test1 test2 ref1 ref2; by seq subj; ilat=0.5*(lat1t+lat2t-lat1r-lat2r); dlat=lat1r-lat2r; run; For some reason SAS won't compute ilat and dlat. Thanks John Edit: Category changed. [Helmut] |
|
Helmut ★★★   Vienna, Austria, 2013-04-09 04:58 (4419 d 01:51 ago) @ jag009 Posting: # 10377 Views: 4,326 |
|
|
Hi John, ❝ I tried it with the EMEA 4-way datasets but I am stuck at the merging routine for "test1 test2 ref1 ref2" and the ilat and dlat calculation. EMA’s data set I is nasty: imbalanced and incomplete. Maybe you have to adjust your merge-command? If you want to compare it to results obtained in Phoenix, see here.— Dif-tor heh smusma 🖖🏼 Довге життя Україна! ![[image]](https://static.bebac.at/pics/Blue_and_yellow_ribbon_UA.png) Helmut Schütz ![[image]](https://static.bebac.at/img/CC by.png) The quality of responses received is directly proportional to the quality of the question asked. 🚮 Science Quotes |
|
jag009 ★★★ NJ, 2013-04-09 22:12 (4418 d 08:37 ago) @ Helmut Posting: # 10385 Views: 4,194 |
|
|
Hi Helmut, ❝ If you want to compare it to results obtained in Phoenix, see here. I figured it out... Here is mine resultsYours from WinnonlinI need to check boundy... Thanks John |
|
Helmut ★★★ Vienna, Austria, 2013-04-10 19:32 (4417 d 11:18 ago) @ jag009 Posting: # 10400 Views: 4,361 |
|
|
Hi John, ❝ I figured it out... And how? Maybe others are interested as well.  ❝ ❝ Theta y s2WR sWR T/R Ratio critBound ❝ 0.79669 -0.15879 0.19931 0.44645 115.461 -0.092076 ❝ ❝ ❝ pointest Title sWR boundy critbound CVwr Assessment ❝ 1.1546132 average 0.44644551 -0.12298597 -0.092076336 46.96 RSABE ❝ ❝ I need to check boundy... Do you get a different value? Can’t imagine since both your y and critbound are the same as mine. χ²0.95,71=91.67 and boundy=y·df/χ² or -0.15879×71/91.67=-0.12299. — Dif-tor heh smusma 🖖🏼 Довге життя Україна! Helmut Schütz The quality of responses received is directly proportional to the quality of the question asked. 🚮 Science Quotes |
|
jag009 ★★★ NJ, 2013-04-10 23:22 (4417 d 07:27 ago) @ Helmut Posting: # 10401 Views: 4,196 |
|
|
Hi Helmut, 1) I made a boo boo on my proc sort and merge command in SAS. That was why the merge didn't work properly. I have to be careful on proc sort and subsequence merge command. 2) My boundy is the same as your boundy (-0.12298597). I had a typo'd on my output. -0.15879 was for y. It's all good. The EMA data is fine. Thanks John |
Ing. Helmut Schütz