interpolating missing values from a TS

Try this (time series with values of 2, NA , etc. at indicated times
(1, 3, etc.) which has NAs removed by linear approx:

library(zoo)
z <- zoo(c(2,NA,1,4,5,2), c(1,3,4,6,7,8))
na.approx(z)

1        3        4        6        7        8
2.000000 1.333333 1.000000 4.000000 5.000000 2.000000

zoo has otherl na.* routines too.  They work the same way.  You just
give them the series and the NAs are filled in as in the example
above.

na.approx - linear approximation to NAs
na.contiguous - only keep portion of series with no NAs
na.locf - fill NAs from last non-NA
na.omit - remove NAs
na.spline - spline approximation to NAs
na.trim - remove NAs at ends

And also in the stinepack package:

na.stinterp - use Stineman approxmation to remove NAs

See ?na.approx for a few more examples.

I've got a large TS with occasional NAs I'd like to roughly
interpolate, I'm currently doing this but it's pretty slow:

? ? ? ?lapply(1:nrow(delta), function(i) {
? ? ? ? ? ? ? ?lapply(1:ncol(delta), function(j) {
? ? ? ? ? ? ? ? ? ? ? ?if (is.na(delta[i,j])) {
? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?if ((i==1)||is.na(delta[i-1,j])) {
? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?delta[i,j] <- delta[i+1,j]
? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?} else { delta[i,j] <- delta[i-1,j]}
? ? ? ? ? ? ? ? ? ? ? ?}
? ? ? ? ? ? ? ?})
? ? ? ?})

any ideas for a better solution?


--
Aleks Clark

_______________________________________________
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.