R help - Nov 2010 - Post-processing of approximated irregular time series

Hi all,

Issue: I merged two zoo objects (a regular and an irregular). After the merge I
used the function 'na.approx' to have also values in the resolution of
the regular time series.

Problem: After approximation some rows at the beginning or at the end of the zoo
objects disappear due to the 'na.approx' algorithm. Now I just want to
have all the rows of the regular time series with the approximated values of the
irregular time series back. Because some rows disappear I wasn't able to
program a code which works properly for 'all' circumstances. I also
tried to work with the 'index' function. Didn't work out so far.

Thank you for your help.

Regards,
Alex

	[[alternative HTML version deleted]]

On Mon, Nov 1, 2010 at 11:14 AM, Salim Alexander (salimale)
<salimale at students.zhaw.ch> wrote:
> Hi all,
>
> Issue: I merged two zoo objects (a regular and an irregular). After the
merge I used the function 'na.approx' to have also values in the
resolution of the regular time series.
>
> Problem: After approximation some rows at the beginning or at the end of
the zoo objects disappear due to the 'na.approx' algorithm. Now I just
want to have all the rows of the regular time series with the approximated
values of the irregular time series back. Because some rows disappear I
wasn't able to program a code which works properly for 'all'
circumstances. I also tried to work with the 'index' function.
Didn't work out so far.
>
> Thank you for your help.
>
> Regards,
> Alex
Note the last line to every message to r-help asking for minimal
reproducible code.   Without that one cannot be sure what the problem
is but note the na.rm= argument on na.approx.zoo .

-- 
Statistics & Software Consulting
GKX Group, GKX Associates Inc.
tel: 1-877-GKX-GROUP
email: ggrothendieck at gmail.com

Hi all,

 

Issue: I merged two zoo objects (a regular and an irregular). After the
merge I used the function 'na.approx' to have also values in the
resolution
of the regular time series.

 

Problem: After approximation some rows at the beginning or at the end of the
zoo objects disappear due to the 'na.approx' algorithm. Now I just want
to
have all the rows of the regular time series with the approximated values of
the irregular time series back. Because some rows disappear I wasn't able to
program a code which works properly for 'all' circumstances. I also
tried to
work with the 'index' function. Didn't work out so far.

 

Thank you for your help.

 

Regards,

Alex

 

***********************************************************

funLm <- function(gauge, orange){

 

gauge <- zoo(gauge$Precipitation, order.by = gauge$Time)

 

tstart <- ISOdatetime(year=2009, month=5, day=1, hour = 0, min = 0, sec = 0,
tz = "UTC")

tend <- ISOdatetime(year=2009, month=5, day=31, hour = 23, min = 59, sec 59,
tz = "UTC")

orangeMai <- window(orange, start=tstart, end=tend) 

gaugeMai <- window(gauge, start=tstart, end=tend)

 

mergeGaugeOrange <- merge.zoo(gaugeMai, orangeMai)

 

smoothing <- na.approx(mergeGaugeOrange)

 

na.index <- which(is.na(mergeGaugeOrange$orangeMai))

na.index <- if(na.index[1] != 1){na.index-(na.index[1]-1)}

 

regularZoo <- smoothing[c(na.index)]

 

rain.lm <- lm(orangeMai ~ gaugeMai, data=regularZoo)

 

par(mfrow=c(2,2))

plot(rain.lm)

 

}

***********************************************************


	[[alternative HTML version deleted]]