Replace NAs by Stineman interpolation
Generic functions for replacing each NA with Stineman interpolated values.
na.stinterp(object, ...) ## Default S3 method: na.stinterp(object, along = time(object), na.rm = TRUE, ...)
object |
object in which |
along |
variable to use for interpolation. Has to be numeric, is otherwise coerced to numeric. |
na.rm |
logical. Should leading and trailing |
... |
further arguments passed to methods. |
Missing values (NAs) are replaced by piecewice rational
interpolation via stinterp.
By default the time index associated with object is used
for interpolation. Note, that if this calls time.default
this gives an equidistant spacing 1:NROW(object). If object
is a matrix or data.frame, the interpolation is done separately for
each column.
An object in which each NA in the input object is replaced
by interpolating the non-NA values before and after it.
Leading and trailing NAs are omitted (if na.rm = TRUE)
or not replaced (if na.rm = FALSE).
Gabor Grothendieck
na.stinterp(c(2,NA,1,4,5,2)) na.stinterp(ts(c(2,NA,1,4,5,2))) ## Not run: #comparison of gap filling with na.stinterp/stinterp and splines, #the solid lines show the continuous interpolation functions #implicitly assumed by the gap-filling, they show how the Stineman #interpolation restricts the range of the interpolant to the #nearby range of the points and suppresses the well known oscillations #characteristic of splines and other methods based on polynomials x <- 1:6 y <- c(2,NA,1,4,5,2) plot(x,y,ylim=c(-1,5)) points(na.stinterp(c(2,NA,1,4,5,2)),pch=2,col=2) points(spline(x,y,n=6),pch=3,col=3) lines(stinterp(x[!is.na(y)],y[!is.na(y)],xout=seq(1,6,by=0.1)),col=2) lines(spline(x,y,n=50),col=3) library(zoo) na.stinterp(zoo(c(2,NA,1,4,5,2))) library(its) na.stinterp(its(c(2,NA,1,4,5,2), seq(Sys.time(), length = 6, by = "day"))) ## End(Not run)
Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.