Date:         Sat, 9 Sep 2006 19:26:45 -0700
Reply-To:     Flamingo <gsuxin@GMAIL.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Flamingo <gsuxin@GMAIL.COM>
Organization: http://groups.google.com
Subject:      Another dataset transformation problem!
Comments: To: sas-l@uga.edu
Content-Type: text/plain; charset="iso-8859-1"

Hi Guys,

Hope everybody has a good weekend.

I have a dataset which is depicted below:

Column 1: Patient ID.
Column 2: ETI. 1 for remote symptomatic etiology; 0 otherwise;
Column 3: EEG. 1 for abnormal electroencephalogram; 0 otherwise;
Column 4: ASL. 1 if the first seizure occured while asleep; 0
otherwise;
Column 5: timel. The time to the 1st seizure recurrence.
Column 6: time2. The time to the 2nd seizure recurrence.


Column 14: timel0. The time to the 10th seizure recurrence.
Column 15: seizl. 1 if the 1st seizure recurrence was observed; 0
otherwise.
Column 16: seiz2. I if the 2nd seizure recurrence was observed; 0
otherwise.


Column 24: seizl0. 1 if the 10th seizure recurrence was observed; 0
otherwise.
We can look at one observation,

20 0 0 0 16.04 41.88 43.83 44.19 54.59 65.41 65.41 65.41 65.41 65.41 1
1 1 1 1 0 0 0 0 0
This patient (ID: 20) experienced five seizure recurrences. From the
initial seizure, the time to
the first seizure recurrence was 16.04, the time to the second
recurrence was 41.88, the time to the third recurrence was 43.83, the
time to the fourth recurrence was 44.19, and the time to the fifth
recurrence was 54.59. The longest follow-up time for this patient was
65.41.
Now we wish to use the time between two successive seizures, instead of
the time from the initial seizureto the ith seizure recurrence. For
example, the above observation should be transformed into six
observations. They are:

                20   0  0  0  16.04  1

                20   0  0  0  25.84  1

                20   0  0  0  1.95    1

                20   0  0  0  0.36    1

                20   0  0  0  10.40   1

                20   0  0  0  10.82   0




We can look at another observation in the original data set.

21 0 1 0 22.53 143.31 143.31 143.31 143.31 143.31 143.31 143.31 143.31
143.31 1 0 0 0 0 0 0 0 0 0

it should be transformed into 2 observations they are

              21  0  1  0  22.53   1

              21  0  1  0  120.78  0

everybody, can you give me some hint. any help is appreciated!