I have time series data with daily dates for each id, I have tried to stset my data using:
Code:
 stset time, id(id) failure(event) scale(365.25)
but the time is incorrectly specified - there are only 2470838 observations representing 1 day each, so a total of 6764 person years... but when I stset the person-time for txgroup==1 is 200380 (but there are 1212256 observations [days] for txgroup==1) which should be 1212256/365.25 = 3318.976 person years...

How do I properly stset my data to get the correct person-time?


Code:
* Example generated by -dataex-. To install: ssc install dataex
clear
input int id float(time txgroup event)
3 19455 1 .
3 19456 1 .
3 19457 1 .
3 19458 1 .
3 19459 1 .
3 19460 1 .
3 19461 1 .
3 19462 1 .
3 19463 1 .
3 19464 1 .
3 19465 1 .
3 19466 1 .
3 19467 1 .
3 19468 1 .
3 19469 1 .
3 19470 1 .
3 19471 1 .
3 19472 1 .
3 19473 1 .
3 19474 1 .
3 19475 1 .
3 19476 1 .
3 19477 1 .
3 19478 1 .
3 19479 1 .
3 19480 1 .
3 19481 1 .
3 19482 1 .
3 19483 1 .
3 19484 1 .
3 19485 1 .
3 19486 1 .
3 19487 1 .
3 19488 1 .
3 19489 1 .
3 19490 1 .
3 19491 1 .
3 19492 1 .
3 19493 1 .
3 19494 1 .
3 19495 1 .
3 19496 1 .
3 19497 1 .
3 19498 1 .
3 19499 1 .
3 19500 1 .
3 19501 1 .
3 19502 1 .
3 19503 1 .
3 19504 1 .
3 19505 1 .
3 19506 1 .
3 19507 1 .
3 19508 1 .
3 19509 1 .
3 19510 1 .
3 19511 1 .
3 19512 1 .
3 19513 1 .
3 19514 1 .
3 19515 1 .
3 19516 1 .
3 19535 6 1
4 20620 1 .
4 20621 1 .
4 20622 1 .
4 20623 1 .
4 20624 1 .
4 20625 1 .
4 20626 1 .
4 20627 1 .
4 20628 1 .
4 20629 1 .
4 20630 1 .
4 20631 1 .
4 20632 1 .
4 20633 1 .
4 20634 1 .
4 20635 1 .
4 20636 1 .
4 20637 1 .
4 20638 1 .
4 20639 1 .
4 20640 1 .
4 20641 1 .
4 20642 1 .
4 20643 1 .
4 20644 1 .
4 20645 1 .
4 20646 1 .
4 20647 1 .
4 20648 1 .
4 20649 1 .
4 20650 1 .
4 20651 1 .
4 20652 1 .
4 20653 1 .
4 20654 1 .
4 20655 1 .
4 20656 1 .
end
format %td time