# st: RE: Rolling Return calculation

 From "Nick Cox" To Subject st: RE: Rolling Return calculation Date Tue, 21 Jun 2005 16:00:38 +0100

```Your example seems ambiguous. Is this
just the product this * last, or
it the cumulative product?

Cumulative products are exp(sum(log()))
so long as all values are positive.

Nick
n.j.cox@durham.ac.uk

kelly johnson

> I have a quick question. Suppose I have data like the following:
>
>
> Date	Return	Return+1	Rolling Ret
> 6/21/2005	0.078	1.078	1.078
> 6/20/2005	0.051	1.051	1.132978
> 6/19/2005	0.04163	1.04163	1.180144
> 6/18/2005	0.025	1.025	1.209647
> 6/17/2005	0.045	1.045	1.264082
> 6/16/2005	0.045	1.045	1.320965
> 6/15/2005	0.01	1.01	1.334175
> 6/14/2005	0.15	1.15	1.534301
> 6/13/2005	0.004	1.004	1.540438
> 6/12/2005	0.05	1.05	1.61746
> 6/11/2005	0.054	1.054	1.704803
> 6/10/2005	0.045	1.045	1.781519
> 6/9/2005	0.016	1.016	1.810024
>
>
> date is the date
> return is a percentage return
> return +1=return+1
>
> how do i create a rolling return in stata? it has a value of
> 1 initially.
> first obs is 1*1.078, the second obs is 1.078*1.1051, and so on.
>
> i have a long data series (1000 obs). how can i code a
> function to generate
> this variable?

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/
```