.-
help for ^stsplit2^                           @net:from http://www.stata.com/users/wgould!(http://www.stata.com/users/wgould)@
.-

Split time-span records
-----------------------

	^stsplit2^ newvarname [^if^ exp]^,^ ^at(^numlist | ^f^ailures^)^ [^nopre^serve ^trim^]

	^stsplit2^ newvarname ^=^ exp [^if^ exp]^,^ ^at(^numlist^)^ [^nopre^serve ^trim^]


^stsplit2^ is for use with survival-time data; see help @st@.
You must ^stset^ your data before using this command; see help @stset@.


Description
-----------

^stsplit2^ is identical to ^stsplit^ (see help @stsplit@) with the addition of one
new feature.

^stsplit2^ allows specification of ^at(failures)^ to indicate records are to
be split at every failure time.


Options for ^stsplit2^
--------------------

^at(^numlist | ^failures)^ is not optional.  It specifies the points at which
    records are to be split.  The points are specified in analysis-time t
    units.

    In ^stsplit2^'s first syntax (no expression specified), ^at(5(5)20)^ would
    split records at t==5, t==10, t==15, and t==20.  Specifying ^at(failures)^
    would split the records at each failure time in the data.

    In ^stsplit2^'s second syntax (expression specified), ^at(5(5)20)^ would split
    the records at the t corresponding to 5, 10, 15, and 20 analysis t units
    after exp, where exp is to evaluate to a time recorded in the same way as
    timevar from "^stset timevar,^ ..."

    In ^stsplit2^'s second syntax, ^at(failures)^ is not allowed.

^nopreserve^ and ^trim^ are as described in ^stsplit^.


Examples
--------

	. ^stsplit2 cat, at(2 4 6 8)^
	. ^stsplit2 cat, at(failures)^


Warning
-------

With large datasets with many failure times, specifying at(failure) can result
in long execution times and results in very large datasets.



Also see
--------

 Manual:  ^[R] st stsplit^
On-line:  help for @stsplit@