IDPmisc: rose – R documentation

Become an expert in R — Interactive courses, Cheat Sheets, certificates and more!

rose

Creates a rose object out of circular data

Description

rose splits data into subsets according to one or two grouping elements, computes summary statistics for each, and returns the result in a rose object.

Usage

rose(x, subset = NULL,
     cyclVar = NULL, circle = NULL, n.cyclVar = 8,
     cut = NULL, labels = NULL,
     breaks = NULL, include.lowest = FALSE, right = TRUE, dig.lab = 2,
     warn = TRUE, FUN = mean, ...)

Arguments

`x`	Vector, data frame or matrix containing the response.
`subset`	An optional vector specifying a subset of observations to be used in the aggregating process.
`cyclVar`	Cyclic variable as first grouping element. `cyclVar` must be a numeric vector whose length is equal to the number of rows in `x` with 0 <= `cyclVar` < `circle`. Observations where cyclic variables are NA are automatically excluded from the rose object.
`circle`	Defines the value of a full circle with no default.
`n.cyclVar`	Defines the number of equally spaced intervals of the cyclic variable, into which the data are split. The first interval is labeled with 0 and is always centered around 0.
`cut`	Vector of numerics, logicals or factors as second grouping elements. Its length is equal to the number of rows in `x`. `cut` is used to group the observations similar to function `cut`.
`labels`	Labels for the corresponding intervals. When cut is a logical, labels has to be named in the order: FALSE, TRUE.
`breaks, include.lowest, right, dig.lab`	These arguments are only active when `cut` is numeric and are used in exactly the same way as in cut. `breaks` defines the break points. `include.lowest = TRUE` indicates that the lowest (or highest, for right = FALSE) breaks value should be also included. `right = TRUE` indicates that the intervals should be closed on the right (and open on the left) or vice versa for `right = FALSE`. `dig.lab` number of digits for breaks labeling when labels are not given explicitely.
`warn`	Logical, indicating if warnings should be issued for NAs in `cyclVar` and / or x-values outside of breaks range.
`FUN`	Summary function, returning a scalar or vector.
`...`	Additional arguments for summary function.

Details

The first grouping element, cyclVar, for the summary statistics must be circular and numeric. The second grouping element, cut, can be numeric, logical or a factor.
Not all combinations of arguments are allowed:
Argument cut can only be defined when summary consists of a scalar and x consists of 1 column.
When x contains only one column and cut is not defined, the summary function may also be a vector with the restriction, that the summary of each subset, defined by the cyclic variable, must have the same number of elements.
When x is a data frame or matrix with more than 1 column, the summary function must be scalar.

Value

Object of class rose

Author(s)

Rene Locher

Examples

## artificial example:
## concentration of A and B as function of
## hour of day (hour) and wind direction (WD)
hour <- rep(0:23,100)
dat <- data.frame(hour = hour,
                  A =    (2*cos((hour+6)/6*pi)+rnorm(24*100,1))^2,
                  B =    (2*cos((hour+4)/6*pi)+rnorm(24*100,1,2))^2,
                  WD =   rnorm(24*100,
                           mean=sample(c(190,220,50),24*100, replace = TRUE),
                           sd=30)%%360)

## two different responses, scalar summary function
mean.windrose <- rose(dat[,c("A","B")],
                      cyclVar=dat$WD,
                      circle=360,
                      FUN=mean, na.rm=TRUE)
mean.windrose

## one response, vectorial summary function
quant.dayrose <- rose(dat$A,
                      cyclVar=dat$hour,
                      n.cyclVar=24, circle=24,
                      FUN=quantile, na.rm=TRUE)
quant.dayrose


mean.windroseB <- rose(dat[,c("A")],
                       cyclVar=dat$WD,
                       circle=360,
                       cut=dat$B,
                       breaks=c(0,30,100),
                       dig.lab=3,
                       FUN=mean, na.rm=TRUE)
mean.windroseB

IDPmisc

'Utilities of Institute of Data Analyses and Process Design (www.zhaw.ch/idp)'

v1.1.20

GPL (>= 3)

Authors

Rene Locher [aut, cre], Andreas Ruckstuhl [ctb]

Initial release

2020-01-21