On Wed, Jun 30, 2010 at 10:05 AM, Research <risk2009 at ath.forthnet.gr>
wrote:> Hello again,
>
> I merge different zoo time series with prices at different dates. This
> returns a multivariate zoo object with NA's at various points i.e.,
>
> 2010-02-28 ? ? NA ? ? NA ? ? ?NA ? ? ? NA ? 850.2 ?2444.4 ? ? NA ? ? ?NA
> ?NA ? ? ?NA ? ? NA ? ? ?NA ? ? ? NA
> 2010-03-01 ? 61.1 ? 55.3 ? ?61.5 ? ?81.24 ? ? ?NA ? ? ?NA 1712.2 ? ? 3.3
> 11139.3 ? 163.7 2242.4 ?9015.6 ?109.791
> 2010-03-31 ? ? NA ? ? NA ? ? ?NA ? ? ? NA ? 846.5 ?2439.0 ? ? NA ? ? ?NA
> ?NA ? ? ?NA ? ? NA ? ? ?NA ? ? ? NA
> 2010-04-01 ? 66.9 ? 49.4 ? ?65.7 ? ?84.48 ? ? ?NA ? ? ?NA 1700.6 ? ? 3.8
> 11203.0 ? 164.4 2245.7 ?9064.5 ?109.887
> 2010-04-30 ? ? NA ? ? NA ? ? ?NA ? ? ? NA ? 838.0 ?2440.0 ? ? NA ? ? ?NA
> ?NA ? ? ?NA ? ? NA ? ? ?NA ? ? ? NA
> 2010-05-01 ? 66.6 ? 45.6 ? ?65.7 ? ?73.84 ? ? ?NA ? ? ?NA 1705.3 ? ? 4.0
> 11252.0 ? 165.1 2248.4 ?9113.9 ?110.065
>
> Will the "fill" parameter of the zoo.merge() allow a missing
value for each
> zoo object merged ?to take the last available (previous in time) price for
> that particular time series or one has to do that using ?the na.locf()
> command afterwards?
Yes, use na.locf after merging. fill is used to fill with a constant
value such as zero.