[R-SIG-Finance] R-Help(Aggregate function): Multiple functions for different collumns

[Joshua Ulrich]

Jatin,

Check out the plyr package.  And there's a quick example of what
you're trying to do on this blog post:
http://www.cerebralmastication.com/?p=339

HTH,
Josh
--
http://www.fosstrading.com



On Wed, Sep 2, 2009 at 11:39 AM, jatin patni<jatinpatni at gmail.com> wrote:
> I'm trying to reduce the resolution or remove duplicate time stamps for the
> same symbols in the following sample file, in a way such that the new price
> is a volume weighted average for that second(timestamp) and the new volume
> is the sum of all volumes within that second.
>
> I'm trying to use sapply and aggregate separately but having trouble with
> the function. I want that the function should do a volume weighted avg for
> the price column in the subgroup(where timestamps and symbols are same) but
> a sum of volume in that group. How do I achieve both goals simultaneously.
>
> I did a lot of research on the mailing archives and spent hours on reading
> the help/faq but couldn't figure it out.
>
> Here's my codes that is working:-
>
> tmp1.split <- sapply(split(tmp1, tmp1$symbol),
> function(x){weighted.mean(tmp1$price, tmp1$volume)})
> tmp1.ag <- aggregate(tmp1, by=list(tmp1$symbol, tmp1$time),
> function(x){weighted.mean(tmp5$price, tmp5$volume)})
>
> How do I go ahead now(I'm confused and new to R)
> How do I write functions that handle multiple columns differently.
>
>>tmp1 #my dataframe
>   sno     symbol series     time   price volume
> 1    1 BHARTIARTL     EQ 18:15:05  600.00      1
> 2    2 BHARTIARTL     EQ 18:15:05  600.00     99
> 3    3   GLENMARK     EQ 18:15:05  238.10      5
> 4    4   HINDALCO     EQ 18:15:05   43.75    100
> 5    5 BHARTIARTL     EQ 18:15:05  600.00      1
> 6    6       BHEL     EQ 18:15:05 1100.00     11
> 7    7   HINDALCO     EQ 18:15:06   43.20      1
> 8    8 CHAMBLFERT     EQ 18:15:06   46.00     10
> 9    9 CHAMBLFERT     EQ 18:15:06   46.00     90
> 10  10 BAJAUTOFIN     EQ 18:15:06   80.00    100
>
> Thanks
>
> --
> Jatin Patni
> Tel: 91 9833 2180 52
> jatinpatni at gmail.com
> www.jatinpatni.co.nr
>
>        [[alternative HTML version deleted]]
>
> _______________________________________________
> R-SIG-Finance at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-finance
> -- Subscriber-posting only.
> -- If you want to post, subscribe first.
>