moleps, looks like your fudate and pfudate in obs 83 are about two
centuries behind. check out the values of those observations without
date formatting, you will see negative integers if i'm right.
generally, it is a good idea to use 4-digit years when formatting dates.
laszlo
moleps islon wrote:
IN generating the timevariabel for a cox model I get the follwing.
Everything is perfect, except id 83. Any idea why??
gen futime=fudate-opdate
gen pfutime=pfudate-opdate
gen coxtime=pfutime if pfutime>futime &pfutime<.
replace coxtime=futime if pfutime==.
replace coxtime=pfutime if futime==.
list opdate fudate pfudate futime pfutime coxtime
| opdate fudate pfudate futime pfutime coxtime |
76. | 10 Nov 04 18 Apr 05 23 May 08 159 1290 1290 |
77. | 25 May 04 16 Sep 04 23 May 08 114 1459 1459 |
78. | 24 Aug 04 01 Dec 04 30 Jun 08 99 1406 1406 |
79. | 28 Feb 05 22 Sep 05 26 Jun 08 206 1214 1214 |
80. | 06 Jan 05 13 Apr 05 15 Oct 08 97 1378 1378 |
|----------------------------------------------------------------|
81. | 10 Jan 05 18 Apr 05 15 Sep 08 98 1344 1344 |
82. | 14 Apr 05 06 Jun 05 25 Jun 08 53 1168 1168 |
83. | 22 Jun 05 04 Oct 05 24 Jun 08 -72944 -71950 -71950 |
84. | 15 Jan 87 10 Feb 96 . 3313 . 3313 |
85. | 31 Aug 92 20 Mar 94 . 566 . 566 |
|----------------------------------------------------------------|
Regards,
Moleps
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
