Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.

# st: Time-span database, stset or snapspan problem

 From "Maria Dolores Montoya Diaz" To Subject st: Time-span database, stset or snapspan problem Date Mon, 3 May 2010 12:14:59 -0300

```Dear all,

My original database has the following firms information:

var1= firm code, var2 = opening date,  var3 = date of the balance sheet of
financial information (quarterly data: trim1/1995 to trim 2/2009),
var4-var20 = financial indicators at different periods, var21= dummy (1=
firm closed, 0= firm in operation),
var22=closing date or date of last financial information available.

Example from two companies(var1==1 and var1==2):

+---------------------------------------------------------------------------
-----+
|   var1    var2        var3          var4  ...   var20       var21
var22 |

|---------------------------------------------------------------------------
-----|
1. |     1   30jun1994   30mar1995    161.1933    .0062037         0
20mar1998 |
2. |     1   30jun1994   30jun1995    255.4849    .0039141         0
20mar1998 |
3. |     1   30jun1994   30sep1995    6.106456    .1637611         0
20mar1998 |
:
:
12. |     1   30jun1994   30dec1997    264.0175    .0037876         1
20mar1998 |
13. |     2   06jun1994   30mar1995    .0717944    13.92866         0
30jun2009 |
14. |     2   06jun1994   30jun1995    .0544895    18.35216         0
30jun2009 |
:
:
70. |     2   06jun1994   30jun2009    .0642468    15.56499         0
30jun2009 |

|---------------------------------------------------------------------------
-----|

The first company closed in 20mar1998 and the last financial information is
from 30dec1997. The second one remains in operation and the var22 is equal
var3 for the last observation available - 30jun2009.

The following Snapspan command(Stata 11) was used:

snapspan var1  var3 var2 var22 var21 var4-var20, generate(data0) replace

After, the stset command:

stset var3, id(var1) origin(time var2) failure(var21==1)

So, the variable survival was calculated for all companies - closed and in
operation - as the date of the last financial information (var3) - (var2)
opening date. I do not have var4-var20 at closing date from closed firms.
How do I get the correct time-span database?  Is it necessary to add an
additional line for closed firms to include in var3 the closing date? In
this case, var4-var20 will be missing? Is it possible to apply snapspan
options to have this solution?

Thanks,
Dolores

____________________________________________________________________________
______
Maria Dolores Montoya Diaz
University of São Paulo - Brazil
FEA-RP/USP - Ribeirão Preto School of Economics, Administration and
Accounting
Departament of Economics
____________________________________________________________________________
______

*
*   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/
```