Prev: Implementing a design for pairwise testing
Next: A Complex marging Issue - Please be kind enough to take a
From: Tom Smith on 9 Mar 2010 14:49
I have a following datatset "Girl" with three variables (SID1A,
AEVSTT1D,AEVEND1D, trt)



SID1A AEVSTT1D AEVEND1D trt

0150_00002 11APR2007 11APR2007 A
0150_00002 18APR2007 20APR2007 A
0150_00002 02MAY2007 04MAY2007 A
0150_00002 05MAY2007 07MAY2007 A
0150_00002 08MAY2007 08MAY2007 A
0150_00002 09MAY2007 10MAY2007 A
0150_00002 11MAY2007 16MAY2007 A
0150_00002 24MAY2007 24MAY2007 A
0150_00002 26MAY2007 26MAY2007 A
0150_00002 27MAY2007 27MAY2007 A
0150_00002 28MAY2007 28MAY2007 A
0150_00002 29MAY2007 29MAY2007 A
0150_00002 30MAY2007 31MAY2007 A
0150_00002 30AUG2008 30MAY2008 A
0012_00002 17NOV2006 23NOV2006 B
0012_00002 03DEC2006 19DEC2006 B
0012_00002 21DEC2006 26DEC2006 B
0012_00002 27DEC2006 19JAN2007 B
0012_00002 27DEC2007 29JAN2008 B




I have another dataset "Boy" as below with three variables
(SID1A,CYCLE,SMDDOS1D)




SID1A CYCLE SMDDOS1D

0150_00002 1 06APR2007
0150_00002 2 27APR2007
0150_00002 3 18MAY2007
0012_00002 1 08NOV2006
0012_00002 2 29NOV2006
0012_00002 3 20DEC2006



I need to merge them (Boy and Girl) such a way that if the value for the
variable
AEVSTT1D, AEVEND1D from the dataset "Girl" falls in between the 2nd value
of SMDDOS1D
and first value of SMDDOS1D from "Boy" (such as anything before 06APR2007
OR such
as (26apr2007 - 06aprl2007).or such as (17MAY2007- 27APR2007)). Also if
last
observation of value for the variable if trt = "A" and AEVEND1D is more
than
28 days after of the value of SMDDOS1D (ONLY the last observation - By
CYCLE,of
the dataset "Boy" )and if trt = "B"and AEVEND1D is more than
21 days after of the value of SMDDOS1D (ONLY the last observation - By
CYCLE, of the dataset
"Boy" ) then they should not be included in teh final dataset.

The result dataset should be as below:



SID1A AEVSTT1D AEVEND1D Cycle

0150_00002 11APR2007 11APR2007 1
0150_00002 18APR2007 20APR2007 1
0150_00002 02MAY2007 04MAY2007 2
0150_00002 05MAY2007 07MAY2007 2
0150_00002 08MAY2007 08MAY2007 2
0150_00002 09MAY2007 10MAY2007 2
0150_00002 11MAY2007 16MAY2007 2
0150_00002 24MAY2007 24MAY2007 3
0150_00002 26MAY2007 26MAY2007 3
0150_00002 27MAY2007 27MAY2007 3
0150_00002 28MAY2007 28MAY2007 3
0150_00002 29MAY2007 29MAY2007 3
0150_00002 30MAY2007 31MAY2007 3
0012_00002 17NOV2006 23NOV2006 1
0012_00002 03DEC2006 19DEC2006 3
0012_00002 21DEC2006 26DEC2006 3
0012_00002 27DEC2006 19JAN2007 3


Tahnks you so much. It's complex and confusing. But if you look the final
result dataset that
make more sense than the description.
 | 
Pages: 1
Prev: Implementing a design for pairwise testing
Next: A Complex marging Issue - Please be kind enough to take a