Re: st: Taking products in loops

[Nick Cox <njcoxstata@gmail.com>]

Naturally that is only a good solution when all values are positive.

Let's take the problem more cautiously. First, code that allows zero
or negative values.

I will assume a -time- or equivalent too.

bysort event (time) : gen double product = return[1]
by event : replace product = product[_n-1] * return if _n > 1

Second, what about missing values? Missing values would usually be
ignored in Stata.

bysort event (time) : gen double product = cond(missing(return[1]), 1,
return[1])
by event : replace product = product[_n-1] * cond(missing(return), 1,
return) if _n > 1

Some might object to 1 being returned as the products of all missings,
which can in turn be fixed

bysort event (return) : replace product = . if missing(return[1]) &
missing(return[_N])
sort event time


On Tue, Mar 27, 2012 at 12:26 AM, Nick Cox <njcoxstata@gmail.com> wrote:
> bysort event : gen double product = exp(sum(ln(return)))
>
> In addition, or rather by way of multiplying solutions, did you remember to
>
> . search product
>
> before you posted? Among various false positives are some pertinent
> reminders that others have had the same problem before you:
>
> STB-60  dm87  . . . . . . . . . .  Calculating the row product of observations
>        (help rprod if installed) . . . . . . . . . . . . . . . . . .  P. Ryan
>        3/01    pp.3--4; STB Reprints Vol 10, pp.39--41
>        generates new variable whose values are the row product of
>        observations
>
> STB-51  dm71  . . . . . . . . . . . .  Calculating the product of observations
>        (help prod if installed)  . . . . . . . . . . . . . . . . . .  P. Ryan
>        9/99    pp.3--4; STB Reprints Vol 9, pp.45--48
>        extension to egen for producing the product of observations
>
> Nick
>
> On Mon, Mar 26, 2012 at 11:44 PM, Sandy Y.  Zhu <sandy.zhu@yale.edu> wrote:
>
>> I was just wondering if you would know how to take products in Stata.
>>
>> Here's how my data looks like:
>>
>> Event            Return
>> 1                    1.1
>> 1                    1.01
>> 1                    0.99
>> 1                    1.23
>> 1                    1.1
>> 1                    1.34
>> 1                    0.88
>> 2                    0.99
>> 2                    0.87
>> 2                    1.1
>> 2                    1.004
>> 3                    1.02
>> 3                    1.09
>> 3                    0.98
>> 3                    0.98
>> 3                    1.09
>> 3                    1.09
>> .....                    ............
>>
>> For each event, I would like to multiply all the numbers under the
>> column "return" for that event together. At the end of the program, I
>> would like to output a column with all the "cumulative returns" for
>> each event. Since each of my events containa  different number of
>> observations (or returns), I'm wondering if there's a command that can
>> do that in 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/