[R] reverse direction with rollapply()

[Joshua Wiley]

That makes sense.  Thanks Gabor!

Josh

On Wed, Jul 21, 2010 at 4:07 PM, Gabor Grothendieck
<ggrothendieck at gmail.com> wrote:
> On Wed, Jul 21, 2010 at 6:59 PM, Joshua Wiley <jwiley.psych at gmail.com> wrote:
>> Hello,
>>
>> I am using rollapply() from package zoo to use a function on unique
>> windows from a dataset.  Because my dataset is not a multiple of my
>> window width, it matters which end I start at.  Here is a simple
>> example, the result I want is 7.5 (i.e., start at the highest level of
>> my ordering variable).  I thought that the argument ascending = FALSE
>> would do it, but it does not seem to have an effect.  It seems like I
>> must be missing something simple.
>>
>> #Library and sample data
>> library(zoo)
>> mydata <- zoo(x = 6:8, order.by = 1:3)
>>
>> rollapply(data = mydata, width = 2, by = 2,
>>  FUN = mean, ascending = TRUE, align = "left")
>>
>> #This gives the same results
>> rollapply(data = mydata, width = 2, by = 2,
>>  FUN = mean, ascending = FALSE, align = "left")
>>
>>
>
> ascending only specifies the order to pass each set of points but does
> not change the sets themselves.  Try this:
>
>
>> rollapply(data = mydata[-1], width = 2, by = 2, FUN = mean)
>  2
> 7.5
>



-- 
Joshua Wiley
Ph.D. Student, Health Psychology
University of California, Los Angeles
http://www.joshuawiley.com/