na.interp1: na.interp1

Description Usage Arguments Value Author(s) Source See Also Examples

Description

This function combines pracma's interp1 constant interpolation method with zoo's na.approx linear interpolation method. Here, x = x rather than x = index(object) in na.approx. Here, y = y rather than y = object in na.approx. Also, here, xi is used instead of xout in na.approx. The Arguments list was obtained from both interp1 and na.approx.

Usage

1
na.interp1(x, y, xi = x, ..., na.rm = TRUE, maxgap = Inf)

Arguments

x

Numeric vector; points on the x-axis; at least two points required; will be sorted if necessary.

y

Numeric vector; values of the assumed underlying function; x and y must be of the same length.

xi

Numeric vector; points at which to compute the interpolation; all points must lie between min(x) and max(x).

...

further arguments passed to methods. The n argument of approx is currently not supported.

na.rm

logical. If the result of the (spline) interpolation still results in NAs, should these be removed?

maxgap

maximum number of consecutive NAs to fill. Any longer gaps will be left unchanged. Note that all methods listed above can accept maxgap as it is ultimately passed to the default method.

Value

Numeric vector representing values at points xi.

Author(s)

Hans Werner Borchers (pracma interp1), Felix Andrews (zoo na.approx), Irucka Embry

Source

  1. zoo's na.approx.R - modified on Fri Aug 6 00:26:22 2010 UTC by felix. See https://r-forge.r-project.org/scm/viewvc.php/pkg/zoo/R/na.approx.R?view=markup&revision=781&root=zoo.

  2. pracma interp1 function definition - R package pracma created and maintained by Hans Werner Borchers. See interp1.

See Also

na.approx, interp1

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
library("iemisc")
library("data.table")

# zoo time series example
zoo1 <- structure(c(1.6, 1.7, 1.7, 1.7, 1.7, 1.7, 1.6, 1.7, 1.7, 1.7,
1.7, 1.7, 2, 2.1, 2.1, NA, NA, 2.1, 2.1, NA, 2.3, NA, 2, 2.1), .Dim = c(12L,
2L), .Dimnames = list(NULL, c("V1", "V2")), index = structure(c(1395242100,
1395243000, 1395243900, 1395244800, 1395245700, 1395256500, 1395257400,
1395258300, 1395259200, 1395260100, 1395261000, 1395261900), class =
c("POSIXct", "POSIXt"), tzone = "GMT"), class = "zoo")

zoo1 <- as.data.frame(zoo1) # to data.frame from zoo
zoo1[, "Time"] <- as.POSIXct(rownames(zoo1)) # create column named Time as a
# POSIXct class
zoo1 <- setDT(zoo1) # create data.table out of data.frame
setcolorder(zoo1, c(3, 1, 2)) # set the column order as the 3rd column
# followed by the 2nd and 1st columns
zoo1 <- setDF(zoo1) # return to data.frame

rowsinterps1 <- which(is.na(zoo1$V2 == TRUE))
# index of rows of zoo1 that have NA (to be interpolated)
xi <- as.numeric(zoo1[which(is.na(zoo1$V2 == TRUE)), 1])
# the Date-Times for V2 to be interpolated in numeric format
interps1 <- na.interp1(as.numeric(zoo1$Time), zoo1$V2, xi = xi,
na.rm = FALSE, maxgap = 1)
# the interpolated values where only gap sizes of 1 are filled
zoo1[rowsinterps1, 3] <- interps1
# replace the NAs in V2 with the interpolated V2 values
zoo1



# data frame time series example
df1 <- structure(list(Time = structure(c(1395242100, 1395243000, 1395243900,
 1395244800, 1395245700, 1395256500, 1395257400, 1395258300, 1395259200,
 1395260100, 1395261000, 1395261900), class = c("POSIXct", "POSIXt"),
 tzone = "GMT"), V1 = c(1.6, 1.7, 1.7, 1.7, 1.7, 1.7, 1.6, 1.7, 1.7, 1.7,
 1.7, 1.7), V2 = c(2, 2.1, 2.1, NA, NA, 2.1, 2.1, NA, 2.3, NA, 2, 2.1)),
 .Names = c("Time", "V1", "V2"), row.names = c(NA, -12L),
 class = "data.frame")

rowsinterps1 <- which(is.na(df1$V2 == TRUE))
# index of rows of df1 that have NA (to be interpolated)
xi <- as.numeric(df1[which(is.na(df1$V2 == TRUE)), 1])
# the Date-Times for V2 to be interpolated in numeric format
interps1 <- na.interp1(as.numeric(df1$Time), df1$V2, xi = xi,
 na.rm = FALSE, maxgap = 1)
# the interpolated values where only gap sizes of 1 are filled
df1[rowsinterps1, 3] <- interps1
# replace the NAs in V2 with the interpolated V2 values
df1



# data.table time series example
dt1 <- structure(list(Time = structure(c(1395242100, 1395243000, 1395243900,
 1395244800, 1395245700, 1395256500, 1395257400, 1395258300, 1395259200,
 1395260100, 1395261000, 1395261900), class = c("POSIXct", "POSIXt"),
 tzone = "GMT"), V1 = c(1.6, 1.7, 1.7, 1.7, 1.7, 1.7, 1.6, 1.7, 1.7, 1.7,
 1.7, 1.7), V2 = c(2, 2.1, 2.1, NA, NA, 2.1, 2.1, NA, 2.3, NA, 2, 2.1)),
 .Names = c("Time", "V1", "V2"), row.names = c(NA, -12L), class =
 c("data.table", "data.frame"), sorted = "Time")

rowsinterps2 <- which(is.na(dt1[, 3, with = FALSE] == TRUE))
# index of rows of x that have NA (to be interpolated)
xi <- as.numeric(dt1[rowsinterps2, Time])
# the Date-Times for V2 to be interpolated in numeric format
interps2 <- dt1[, na.interp1(as.numeric(Time), V2, xi = xi,
 na.rm = FALSE, maxgap = 1)]
# the interpolated values where only gap sizes of 1 are filled
dt1[rowsinterps2, `:=` (V2 = interps2)]
# replace the NAs in V2 with the interpolated V2 values
dt1

iemisc documentation built on Aug. 2, 2020, 9:07 a.m.