Re: st: cumulative summation

[Nick Cox <njcoxstata@gmail.com>]

It works, but does not do what you want. Many solutions, one of which is

egen nties = max(ties), by(group year)

See also -help collapse- on your last.

Nick

On Tue, Feb 26, 2013 at 8:39 AM, James Bernard <jamesstatalist@gmail.com> wrote:
> Thanks Nick,
>
> I have missing values in my list, so it seems that the command is not
> working, because it sets many of the values to . (missing).
>
> I gave you the simplifies version. In fact, I am using loops and
> macros to do this
>
> Then, instead would it be possible to write a comand that can detect
> that for firm A in 1990, 2 is the larger one.
>
> Alternatively, would it be possible to ask the program to keep only
> the last observation of every group (a group defined by firm and year)
>
> Thanks a gain,
> James
>
>
> On Tue, Feb 26, 2013 at 3:49 PM, Nick Cox <njcoxstata@gmail.com> wrote:
>> bysort firm year (ties) : replace ties = ties[_N]
>>
>> Nick
>>
>> On Tue, Feb 26, 2013 at 7:08 AM, James Bernard <jamesstatalist@gmail.com> wrote:
>>
>>> I have used the sum function to calculate a variable in  a panel data.
>>>
>>> Firm        Year      ties
>>> -----------  ----------    -------
>>> A             1990      1
>>> A             1990       2
>>> A             1991       1
>>> B             1990       0
>>> B             1990       1
>>> B             1990        3
>>> C             1990        0
>>> C             1991        0
>>> C             1992        0
>>>
>>> The issue is that the variable ties which is the number of ties a firm
>>> has formed in a year is a cumulative sum of something and, therefore,
>>> for a company like A the final number of ties in year 1990 should be
>>> 2. I need the program to write the same number (final value of the
>>> cumulative sum) instead of 1.
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/