Become an expert in R — Interactive courses, Cheat Sheets, certificates and more!
Get Started for Free

na

replace NA attribute values; disaggregation time series


Description

replace NA attribute values in time series, using last or next observation, or using (temporal) interpolation, and disaggregation

Usage

## S3 method for class 'STFDF'
na.locf(object, na.rm = FALSE, ...)
## S3 method for class 'STFDF'
na.approx(object, x = time(object), xout, ..., na.rm = TRUE)
## S3 method for class 'STFDF'
na.spline(object, x = time(object), xout, ..., na.rm = TRUE)

Arguments

object

object of class STFDF, with potentially NA values

na.rm

logical; need non-replaced NA values be removed?

x

times at which observations are taken; should not be modified

xout

if present, new times at which the time series should be approximated (disaggregated)

...

passed on to underlying zoo functions; see details

Details

details are found in na.locf, na.approx, na.spline.

Value

object of class STFDF, with NA values replaced.

Author(s)

Edzer Pebesma

References

https://www.jstatsoft.org/v51/i07/

Examples

# toy example:
library(sp)
pts = SpatialPoints(cbind(c(0,1),c(0,1)))
Sys.setenv(TZ="GMT")
tm = seq(as.POSIXct("2012-11-25"), as.POSIXct("2012-11-30"), "1 day")
df = data.frame(a = c(NA,NA,2,3,NA,NA,NA,2,NA,NA,4,NA), b = c(NA,2,3,4,5,1,2,NA,NA,NA,NA,3))
x = STFDF(pts, tm, df)
as(x, "xts")
as(na.locf(x), "xts")
as(na.locf(x, fromLast = TRUE), "xts")
as(na.locf(na.locf(x), fromLast = TRUE), "xts")
# drops first record:
as(na.approx(x[,,1]), "xts")
# keep it:
cbind(as(na.approx(x[,,1], na.rm=FALSE), "xts"),
as(na.approx(x[,,2]), "xts"))
cbind(as(na.spline(x[,,1]), "xts"),
as(na.spline(x[,,2]), "xts"))
#disaggregate:
xout = seq(start(x), end(x), "6 hours")
as(na.approx(x[,,1], xout = xout), "xts")
as(na.spline(x[,,1], xout = xout), "xts")
as(na.spline(x[,,2], xout = xout), "xts")

# larger/real data:
data(air)
rural = STFDF(stations, dates, data.frame(PM10 = as.vector(air)))
# fill NA's with last non-NA
r = na.locf(rural)
# sample (NOT aggregate) to monthly:
m = seq(start(rural), end(rural), "1 month")
stplot(na.approx(rural[1:20,"2003::2005"], xout = m), mode = 'ts')

spacetime

Classes and Methods for Spatio-Temporal Data

v1.2-4
GPL (>= 2)
Authors
Edzer Pebesma [aut, cre] (<https://orcid.org/0000-0001-8049-7069>), Benedikt Graeler [ctb], Tom Gottfried [ctb], Robert J. Hijmans [ctb]
Initial release

We don't support your browser anymore

Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.