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

Description

Different high-level graphics functions for displaying large datasets, displaying circular data in a very flexible way, finding local maxima, brewing color ramps, drawing nice arrows, zooming 2D-plots, creating figures with differently colored margin and plot region. In addition, the package contains auxiliary functions for data manipulation like omitting observations with irregular values or selecting data by logical vectors, which include NAs. Other functions are especially useful in spectroscopy and analyses of environmental data: robust baseline fitting, finding peaks in spectra, converting humidity measures.

Author(s)

Rene Locher, Andreas Ruckstuhl et al. Maintainer: Christoph Hofer <christoph.hofer@zhaw.ch>

Pretty Open or Closed Arrows

Description

Draws a set of open or closed arrows which can be shaped by many arguments. Arrows is an extended version of p.arrows.

Usage

Arrows(x1, y1, x2, y2, size = 1, width = 1.2/4/cin, open = TRUE,
       sh.adj = 0.1, sh.lwd = 1, sh.col = par("fg"),
       sh.lty = 1,
       h.col = sh.col, h.col.bo = sh.col, h.lwd = sh.lwd, h.lty = sh.lty,
       verbose = FALSE)

Arguments

Details

Definition of sh.adj:

=0

End of shaft at top of head,

=1

End of shaft at bottom of head,

>1

Gap between shaft and head,

<0

Head is on the shaft.

This function is based on graphics

Value

A value is only returned, when verbose == TRUE.

Note

The plotting device should not be resized manually after plotting as this changes in general the aspect ratio of the plot and deforms hereby the plotted arrows. The beauty of the arrows depends from the resolution of the device. The higher resolutions give better results.

Author(s)

Andreas Ruckstuhl, refined by Rene Locher

See Also

p.arrows, arrows

Examples

## a bunch of different arrows
plot(c(0,10), c(0,10), type="n")
Arrows(5, 5, 5,10, size=3,
       sh.lwd=5, sh.lty=2,
       h.lwd=5)
Arrows(5, 5, 7.5, 9, size=3, open=TRUE,
       sh.adj=0.7, sh.lwd=5, sh.lty=2,
       h.col.bo="red",h.lwd=5)
Arrows(5, 5, 9, 7.5, size=3, open=FALSE,
       sh.adj=1, sh.lwd=5, sh.col="blue",
       h.col.bo="red",h.lwd=2)
Arrows(5, 5, 10, 5, size=2.5, width=1.5, open=FALSE,
       sh.adj=1, sh.lwd=7, sh.col="blue")
Arrows(5, 5, 9, 2.5, size=4, open=FALSE,
       sh.lty=0,
       h.col.bo="black",h.lwd=5)
Arrows(5, 5, 7.5, 1)
Arrows(5, 5, 5, 0, size=2)
Arrows(5, 5, 2.5, 1, size=2, width=1)

## vector field
x<- runif( 20)
y<- runif( 20)
u<- 0.1+0.02*rnorm(20)
v<- 0.1+0.02*rnorm(20)
plot(x,y,xlim=range(c(x,x+u)),ylim=range(c(y,y+v)),type="n")
Arrows(x,y,x+u,y+v,sh.col="blue")

Color Ramp for Ordered Values

Description

Produces color ramps which change simultanously hues, saturation and values as defined in the hsv modus. This allows to produce especially smooth transitions from one color to the next. The default color ramp starts with light blue, continues with green, yellow, red and ends with dark violet.

Usage

IDPcolorRamp(n,
             colInt = data.frame(h = c(0.47, 0.28, 0.16, 0, 1, 0.8),
                                 s = c(0.31, 0.55, 0.7, 0.8, 0.8, 1),
                                 v = c(1, 1, 1, 1, 1, 0.4)),
             fr = c(0.27, 0.27, 0.27, 0))

Arguments

Details

The function distributes the number of colors in the subramps, given the fractions fr, as smoothly as possible. The default arguments are optimized to most distinct colors possible, also for very small n. There is at least one color in the first and the last subramp.
Definition of hsv code:

h

Hue of hsv-Signal: 0=red, 1/3=green, 2/3=blue, 1=red.

s

Saturation of hsv-Signal: 0=white, 1=full color.

v

Value of hsv-Signal: 0=black, 1=full color.

Value

A vector of n colors.

Note

If there are subramps which are not adjacent in the color space (as is here the case for red and violet), you need a virtual subramp (here from h=0.00 to h=1.00) with corresponding fraction fr == 0.

Author(s)

Rene Locher

See Also

showColors, ColorBrewer

Examples

IDPcolorRamp(10)

## Default IDPcolorRamp in 21 colors
n <- 21
showColors(IDPcolorRamp(n),border=FALSE)

## colorRamp optimized to return at equidistant indices the colors
## light blue, light green, yellow, orange, red, dark violet
## works fine with n > 7
cInt <- data.frame(h = c(0.47, 0.28, 0.16, 0, 1, 0.8),
                   s = c(0.31, 0.55, 0.7, 0.8, 0.8, 1),
                   v = c(1, 1, 1, 1, 1, 0.5))

fr <- c(0.15, 0.25, 0.45, 0.0)
ii <- seq(1,n,length.out=6)

## colors at equidistant indices
showColors(IDPcolorRamp(n, colInt = cInt, fr =fr)[ii], border=FALSE)

## Alternative ramp in 21 colors
showColors(IDPcolorRamp(n, colInt = cInt, fr =fr), border=FALSE)

Display the Density of Points in a Scatter Plot by Colors

Description

The density of points in a scatter plot is encoded by color.

Usage

Image(x, y = NULL, pixs = 1, zmax = NULL, ztransf = function(x){x},
      colramp = IDPcolorRamp, factors = c(FALSE, FALSE),
      matrix = FALSE)

Arguments

Details

Before calling Image a plot must have been created by, e.g., calling plot(x,y,type="n"). This function ensures by default that the pixel has the same size in x- and y-direction. As a drawback, pixels may be unequally spaced, when there are only very few distinct (integer) values in x- or y-direction. When this is the case, the corresponding dimension should be declared as a factor. (cf. argument factors).

This function is based on graphics

Value

Maximum number of counts per pixel found (matrix = FALSE) or the full matrix.

Author(s)

Andreas Ruckstuhl, Rene Locher

See Also

ipairs, ilagplot, iplot, image

Examples

plot.default(iris$Species, iris$Petal.Width, xlim=c(0.5,3.5),
             type="n", axes=FALSE)
axis(1, at=1:3, labels=levels(iris$Species))
axis(2)
Image(iris$Species, iris$Petal.Width, pixs=3)

##
x <- rnorm(10000)
y <- rnorm(10000, 10)
plot(x+y, y, type="n")
Image(x+y, y)
abline(a=0, b=1)

## The above can be merged to
iplot(x+y, y, legend=FALSE, oma=c(5, 4, 4, 2) + 0.1)
abline(a=0, b=1)

Spectrum Measured by a SELDI TOF Mass Spectrometer

Description

The spectrum was taken from a sample of sheep blood. The instrument used was a so called SELDI TOF (Surface Enhanced Laser Desorption Ionisation, Time Of Flight) Mass Spectrometer.

Usage

data(MS)

Format

A data frame with 45213 observations on the following 2 variables.

mz

mass / charge

I

Intensity

Details

The measured masses lie between m/z=1000 and 200000. The intensities are raw output. Neither smoothing nor background subtraction was applied to the spectrum.

Source

Medical research project.

Examples

data(MS)

MS1 <- log10(MS[MS$mz>12000&MS$mz<1e5,])
P <- peaks(MS1, minPH=0.025, minPW=0.0015)

plot(MS1, ty="l", xlim=c(4.15,5))
points(P,col="red")

Omit Observations with NA, NaN, Inf and -Inf Values

Description

Omits observations with values which a not regular (=Not a Regular Value) when object is a vector, a factor, a data.frame or a matrix.

Usage

NaRV.omit(x)

Arguments

Details

Irregular values are defined as NA, NaN, Inf and -Inf Values in numerics and NA in factors and characters.

Value

Returns objects of class vector, factor, data.frame or matrix in the same way as na.omit does. Returns all other objects unchanged and prints a warning.

Author(s)

Rene Locher

See Also

na.omit

Examples

M <- matrix(c(NA,1:7,NA),nrow=3)
M
NaRV.omit(M)

DF <- iris[sample(1:nrow(iris),12),]
DF[1,1] <- NA
DF[10,5] <- NA
row.names(DF) <- 1:12
DF
NaRV.omit(DF)

NaRV.omit(c(NA,1:10,NA))

fac <- factor(c(NA,sample(c(1:9))))
NaRV.omit(fac)

fac <- factor(c(NA,sample(c(1:9))),exclude=NULL)
fac
NaRV.omit(fac)

Convert Cartesian Coordinates to Clock Coordinates

Description

Converts cartesian coordinates (x, y to clock coordinates (rho, phi)

Usage

cart2clock(x, y, circle)

Arguments

Details

Be aware that in clock coordinates and polar coordinate the angle phi is differently defined!

Value

Data frame with

Author(s)

Rene Locher

See Also

clock2cart, rose-class

Examples

## convert clock coordinates to cartesian coordinates
xy <- clock2cart(rho=rep(1,33),phi=seq(0,to=360,length.out=33),circle=360)

## convert the cartesian coordinates back to clock coordinates
rhophi <- cart2clock(xy$x,xy$y,circle=360)

round(clock2cart(rhophi,circle=360)-xy)
## QED

Convert Clock Coordinates to Cartesian Coordinates

Description

Converts clock coordinates (rho, phi) to cartesian coordinates (x, y).

Usage

clock2cart(rho, phi, circle)

Arguments

Details

Be aware that in clock coordinates and polar coordinate the angle phi is differently defined!

Value

Data frame with

Author(s)

Rene Locher

See Also

cart2clock, rose-class

Examples

## an easy way to plot a circle
xy <- clock2cart(rho=rep(1,33),phi=seq(0,to=360,length.out=33),circle=360)
plot(xy)

Convert Color to hsv Code

Description

Converts color names or values to hsv code.

Usage

col2hsv(col)

Arguments

Value

Matrix with the rows

Author(s)

Rene Locher

See Also

col2rgb, rgb2hsv, hsv

Examples

col2hsv(c("blue","green","red"))

Coerce a list to a data.frame

Description

Coerces a list with vectors of different length into a data.frame. Fills the shorter vectors with NA.

Usage

data.sheet(x)

Arguments

Details

This function is convenient for comparing / controling data in Lists whose components should have approximately the same length.

Value

data.frame

Author(s)

Thomas Unternaehrer

Examples

data.sheet(list(a=1:5,b=1:4))

Produce a Legend or Key (Grid Function)

Description

Produces (and possibly draws) a Grid frame grob which is a legend that can be placed in other Grid plots. draw.leg is a slightly enhanced version of draw.key in package lattice V 0.12-3.

Usage

draw.leg(key, draw=FALSE, vp=NULL)

Arguments

Details

Three new components are added to the list key of the original code in package lattice V 0.12-3: between.rows, between.title, adj.title. They allow to format the legend in a more flexible way. To ease the use of draw.leg, the full description of draw.key and the here interesting part of xyplot are also included:
The key essentially consists of a number of columns, possibly divided into blocks, each containing some rows. The contents of the key are determined by (possibly repeated) components named “rectangles”, “lines”, “points” or “text”. Each of these must be lists with relevant graphical parameters (see later) controlling their appearance. The key list itself can contain graphical parameters, these would be used if relevant graphical components are omitted from the other components.

The length (number of rows) of each such column (except “text”s) is taken to be the largest of the lengths of the graphical components, including the ones specified outside (see the entry for rep below for details on this). The “text” component has to have a character or expression vector as its first component, and the length of this vector determines the number of rows.

The graphical components that can be included in key (and also in the components named “text”, “lines”, “points” and “rectangles” as appropriate) are:

• cex=1

• col="black"

• lty=1

• lwd=1

• font=1

• fontface

• fontfamily

• pch=8

• adj=0

• type="l"

• size=5

• angle=0

• density=-1

adj, angle and density are currently unimplemented. size determines the width of columns of rectangles and lines in character widths. type is relevant for lines; "l" denotes a line, "p" denotes a point, and "b" and "o" both denote both together.

Other possible components of key are:

between

numeric vector giving the amount of space (character widths) surrounding each column (split equally on both sides),

title

string or expression giving a title for the key

rep

logical, defaults to TRUE. By default, it's assumed that all columns in the key (except the “text”s) will have the same number of rows, and all components are replicated to be as long as the longest. This can be suppressed by specifying rep=FALSE, in which case the length of each column will be determined by components of that column alone.

cex.title

cex for the title

background

background color, defaults to default background

border

either a color for the border, or a logical. In the latter case, the border color is black if border is TRUE, and no border is drawn if it is FALSE (the default)

transparent=FALSE

logical, whether key area should have a transparent background

columns

the number of columns column-blocks the key is to be divided into, which are drawn side by side.

between.columns

Space between column blocks, in addition to between.

between.rows

New argument: Space between rows.

between.title

New argument: Space between top row and title of legend.

adj.title

New argument: Adjustment of title in respect to body of legend.
adj.title = 0 adjusts the title to the left,
adj.title = 1 adjusts the title to the right and
adj.title = 0.5 (=default) centers the title.

divide

Number of point symbols to divide each line when type is "b" or "o" in lines.

Value

A Grid frame object (that inherits from “grob”)

Author(s)

Deepayan Sarkar, modified by Rene Locher

See Also

xyplot

Examples

require(grid)

## Legend 2 cm below and 1 cm left of the upper right corner
grid.newpage()
key <- draw.leg(key =
                    list(rectangles = list(col=1:3,
                             size = 2,
                             lwd = 0.5),
                         text = list(LETTERS[1:3]),
                         cex = 1.2,
                         between = 2,
                         between.rows = 0.5,
                         between.title = 0.7,
                         title = "component",
                         cex.title = 1.4,
                         transparent = TRUE))

vp.key <-  viewport(x = convertX(unit(1,"npc")-unit(1,"cm"),"cm"),
                    y = convertY(unit(1,"npc")-unit(2,"cm"),"cm"),
                    width=grobWidth(key),
                    height=grobHeight(key),
                    just=c("right","top"))

pushViewport(vp.key)
grid.draw(key)
popViewport()

##----------------------------------------
## Legend 1 cm above the lower left corner
key <- draw.leg(key = list(rectangles = list(col=1:3,
                             size = 4,
                             lwd = 0.5),
                  text=list(letters[1:3]),
                  lines=list(col=1:3),
                  cex=2,
                  between=2,
                  between.rows=0.5,
                  between.title=0.7,
                  title = "component",
                  adj.title = 0,
                  cex.title = 2.4,
                  transparent = TRUE))

vp.key <-  viewport(x = 0,
                    y = unit(1,"cm"),
                    width=grobWidth(key),
                    height=grobHeight(key),
                    just=c("left","bottom"))

grid.newpage()
pushViewport(vp.key)
grid.draw(key)
popViewport()

Auxilary for Controlling the General Appearance of a Rose Plot

Description

Defines the General Appearance of rose plot.

Usage

general.control(stacked = FALSE,
                rose.rad = NULL, rose.x = NULL, rose.y = NULL,
                mar = rep(0.3, 4),
                rev.col = FALSE,
                shift = 0,
                cex = 1, col = NULL, lty = 1:3, lwd = 1, type = "s")

Arguments

Value

Returns the arguments conveniently packaged up in a list to supply the general arguments for plot.rose.

Author(s)

Rene Locher

See Also

plot.rose, grid.control

Examples

general.control()

Easy and Flexible Input for One- and Two Dimensional Data

Description

The function accepts one- or two dimensional data, checks them for compatibility and gives a dataframe back.

Usage

getXY(x, y = NULL, unidim.allowed = TRUE)

Arguments

Details

Input is checked for compatibility: When x is a list, the first two elements must have identical length. When y is defined, x must be one dimensional and must have the same length as argument y. When onedim.allowed is FALSE, input must be twodimensional.

When input is one dimensional and unidim is TRUE, y gives the numbers of elements back.

Function works similar to xy.coords

Value

dataframe with the components x and y

Author(s)

Rene Locher

Examples

getXY(3:4, 1:2)
getXY(matrix(1:4,ncol=2))
getXY(as.data.frame(matrix(1:4,ncol=2)))
getXY(4:1)
getXY(list(a=1:2,b=9:10))

Auxilary for Controlling the Grid Appearance of a Rose Plot

Description

Defines the appearance of the guiding elements of rose plots such as circles, rays and labels.

Usage

grid.control(circ.n = 4, circ.r = NULL,
             circ.col = "gray30", circ.lwd = 0.5,
             circ.cex = 0.8, circ.between = 0.3,
             circ.dir = pi/16 * 9,
             circ.sub.n = NULL, circ.sub.r = NULL,
             circ.sub.col = "gray70", circ.sub.lwd = 0.5,
             cyclVar.lab = c("N","NE","E","SE","S","SW","W","NW"),
             cyclVar.cex = 1.2, cyclVar.between = 0,
             cyclVar.centered = TRUE,
             ray.lim = NULL, ray.n = 8)

Arguments

Value

Returns the arguments conveniently packaged up in a list to supply the arguments for the grid appearance of rose plot.

Author(s)

Rene Locher

See Also

plot.rose, general.control

Examples

grid.control()

Converting Humidity Measures into Each Other

Description

Converting dew point (of water in air) into vapor pressure or relative humidity and vice versa.

Usage

hr(T, Td, warn = TRUE)
pw(Td, warn = TRUE)
pw.ai(Td, warn = TRUE)
pw.aw(Td, warn = TRUE)
Td(pw = NULL, T = NULL, hr = NULL, warn = TRUE)
Td.aw(pw = NULL, T = NULL, hr = NULL, warn = TRUE)
Tf.ai(pw = NULL, T = NULL, hr = NULL, warn = TRUE)

Arguments

Details

All vapour pressures correspond to pure water vapour and are not adjusted to water vapour in air.

Value

hr transforms dew point above ice / water at a certain temperature into relative humidity (-65^\circ\textnormal{C} <= T <= 60^\circ\textnormal{C}).
pw transforms dew point into saturation vapour pressure above ice (-65^\circ\textnormal{C} <= T <= 0^\circ\textnormal{C}) and above water (0^\circ\textnormal{C} < T <= 60^\circ\textnormal{C}) respectively.
pw.ai transforms dew point into saturation vapour pressure above ice (-65^\circ\textnormal{C} <= T <= 0^\circ\textnormal{C}).
pw.aw transforms dew point into saturation vapour pressure above liquid water (-45^\circ\textnormal{C} <= T <= 60^\circ\textnormal{C}).
Td transforms vapour pressure or relative humidity and temperature into frost point above ice (-65^\circ\textnormal{C} <= T <= 0^\circ\textnormal{C}) and dew point above water (0^\circ\textnormal{C} < T <= 60^\circ\textnormal{C}) respectively.
Td.aw transforms vapour pressure or relative humidity and temperature into dew point above water (-45^\circ\textnormal{C} <= T <= 60^\circ\textnormal{C}).
Tf.ai transforms vapour pressure or relative humidity and temperature into frost point above ice (-65^\circ\textnormal{C} <= T <= 0^\circ\textnormal{C}).

Author(s)

Rene Locher

References

Guide to Meteorological Instruments and Methods of Observation, WMO, WMO-No. 8, Seventh edition, 2008, updated 2010, 1.4-29 Annex 4.B

Examples

 pw(Td = c(-65, -45, 0, 30, 60))
 Td(pw = pw(c(-20, 0, 20)))

 hr(T = c(20, 30, 40), Td = c(0, 20, 30))
 ## [1] 26.20257 55.09561 57.46519

 hr(T = seq(0, 60, 5), Td = 0)

Image Lag Plot Matrix for Large Time Series

Description

Produces an image lag plot matrix of large timeseries where the colors encode the density of the points in the lag plots.

Usage

ilagplot(x, set.lags = 1,
         pixs = 1, zmax = NULL, ztransf = function(x){x},
         colramp = IDPcolorRamp, mfrow=NULL, cex=par("cex"),
         main = NULL, d.main = 1, cex.main = 1.5*par("cex.main"),
         legend = TRUE, d.legend = 1,
         cex.axis = par("cex.axis"), las = 1,
         border=FALSE, mar = c(2,2,2,0), oma = rep(0,4)+0.1,
         mgp = c(2,0.5,0)*cex.axis, tcl = -0.3, ...)

Arguments

Details

Code is based on R function lag.plot V1.7. Tip: Legend looks better when mar is defined symmetrically.

Value

Maximum number of counts per Pixel found.

Note

When you get the error message "Zmax too small! Densiest aereas are out of range!" you must run the function with identical parameters but without specifying zmax. The value returned gives you the minimum value allowed for zmax.

Author(s)

Andreas Ruckstuhl, refined by Rene Locher

See Also

ipairs, iplot, Image

Examples

if(require(SwissAir)) {
  data(AirQual)

  ilagplot(AirQual[,c("ad.O3")],set.lags = 1:4,
           ztransf = function(x){x[x<1] <- 1; log2(x)},
           main = "Low correlation")

  Ox <- AirQual[,c("ad.O3","lu.O3","sz.O3")]+
    AirQual[,c("ad.NOx","lu.NOx","sz.NOx")]-
      AirQual[,c("ad.NO","lu.NO","sz.NO")]
  names(Ox) <- c("ad","lu","sz")
  ilagplot(Ox$ad,set.lags = 1:4,
           ztransf = function(x){x[x<1] <- 1; log2(x)},
           main = "High correlation")

  ## cf. ?AirQual for the explanation of the physical
  ## and chemical background
} else print("Package SwissAir is not available")

Internal Functions

Description

These functions are for internal use only and not documented further.

Usage

iplotMain(main, cex.main, cex)

plotPage(y1, y2, names1, names2,
         startP, upf, fpp, overlap,
         x.at, x.ann, x.tick,
         y1.at, y1.ann, y1.tick,
         y2.at, y2.ann, y2.tick,
         ny.ann, cex.ann, xlab, y1lab, y2lab, las,
         col.y1, col.y2,
         cex.lab, y1lim, y2lim, lty1, lty2, lwd1, lwd2, col1, col2,
         leg, y1nam.leg, y2nam.leg, ncol.leg, cex.leg = 1.5,
         h1, h2, col.h1, col.h2,
         mgp, main, cex.main, xpd, cex,
         type1, type2, pch1, pch2, cex.pt1, cex.pt2)

rose.grob(rho, cyclVar, circle, transf, general, grid, title, gdat)

segments.circle(rho, shift, circle = 360, ncp = 1000, gp = gpar())

griddat(rho, cyclVar, circle, vp, grid, title)

plotdat(rho, cyclVar, circle, transf, general, grid)

Details

Author(s)

Rene Locher

Image Scatter Plot Matrix for Large Datasets

Description

Produces an image scatter plot matrix of large datasets where the colors encode the density of the points in the scatter plots.

Usage

ipairs(x,
       pixs = 1, zmax = NULL, ztransf=function(x){x},
       colramp = IDPcolorRamp, cex = par("cex"),
       lab.diag, cex.diag = NULL,
       main = NULL, d.main = 1.5, cex.main = 1.5*par("cex.main"),
       legend = TRUE, d.legend = 1.5, cex.axis = 0.8*par("cex.axis"),
       nlab.axis = 5, minL.axis = 2, las = 1, border = FALSE,
       mar = rep(0,4), oma = c(3,3,1,0), mgp = c(2,0.5,0)*cex.axis,
       tcl = -0.3, ...)

Arguments

Details

The idea is similar to gplot.hexbin. The hexagons are better suited to reflect the density of points in a plane than the squares used here. Nevertheless squares are, contrary to hexagons, invariant to reflexions at the x- and y-axis and therefore better suited for scatter plot matrices and also for plotting factors.

The code is based on R function pairs V1.7.

Value

Maximum number of counts per Pixel found. Additional elements are returned when verbose == TRUE.

Note

When you get the error message "Zmax too small! Densiest aereas are out of range!" you must run the function with identical parameters but without specifying zmax. The value returned gives you the minimum value allowed for zmax.

Author(s)

Andreas Ruckstuhl, Rene Locher

See Also

ilagplot, iplot, Image

Examples

## Small numbers of different values are plotted nicer
## when converted to factors
AQ <- airquality
AQ$Month <- as.factor(AQ$Month)

zmax <- ipairs(AQ, pixs=2, main="Air Quality")
ipairs(AQ, pixs=2, zmax=zmax, main="Air Quality",border=TRUE)

## example with factors
ipairs(iris,pixs=2)

## a really huge dataset
## Not run: 
  if(require(SwissAir)) {
    data(AirQual)

    ## low correlation
    dev.new()
    ipairs(AirQual[,c("ad.O3","lu.O3","sz.O3")],
    ztransf=function(x){x[x<1] <- 1; log2(x)})
    
    dev.new()
    ipairs(AirQual[,c("ad.NO","lu.NO","sz.NO")],
    ztransf=function(x){x[x<1] <- 1; log2(x)})
    
    ## high correlation
    Ox <- AirQual[,c("ad.O3","lu.O3","sz.O3")]+
    AirQual[,c("ad.NOx","lu.NOx","sz.NOx")]-
    AirQual[,c("ad.NO","lu.NO","sz.NO")]
    names(Ox) <- c("ad","lu","sz")
    dev.new()
    ipairs(Ox, ztransf=function(x){x[x<1] <- 1; log2(x)})

    ## cf. ?AirQual for the explanation of the physical and
    ## chemical background
  } else print("Package SwissAir is not available")

## End(Not run)

Panelplot for itermplot

Description

An example of a useful panel function for huge datasets

Usage

ipanel.smooth(x, y = NULL, pixs = 1, zmax = NULL,
              ztransf = function(x) {x},
              colramp = IDPcolorRamp, col = "black", lwd = 2,
              span = 2/3, iter = 3, ...)

Arguments

Author(s)

Rene Locher

See Also

lowess

Examples

r.lm <- lm(Sepal.Length~Sepal.Width+Petal.Length+Petal.Width+Species,
           data = iris)
par(mfrow = c(2,2), pty = "s")
itermplot(r.lm, se = TRUE, partial.res = TRUE, smooth = ipanel.smooth,
          lwd.smth = 3, pixs = 2, ask = FALSE)

if (require(SwissAir)) {
  par(mfrow = c(1,1))
  dat <- log(AirQual[,c("ad.O3", "ad.WS")])
  pairs(dat,
        panel = ipanel.smooth)
} else print("Package SwissAir is not available")

Image Scatter Plot for Large Datasets

Description

Produces an image scatter plot of large datasets where the colors encode the density of the points in the scatter plot. Works also with factors.

Usage

iplot(x, y = NULL,
      pixs = 1, zmax = NULL, ztransf = function(x){x},
      colramp = IDPcolorRamp, cex = par("cex"),
      main = NULL, d.main = 1, cex.main = par("cex.main"),
      xlab = NULL, ylab = NULL, cex.lab = 1,
      legend = TRUE, d.legend = 1,
      cex.axis = par("cex.axis"), nlab.xaxis = 5, nlab.yaxis = 5,
      minL.axis = 3, las = 1, border = FALSE,
      oma = c(5,4,1,0)+0.1, mgp = c(2,0.5,0)*cex.axis, tcl = -0.3, ...
      )

Arguments

Details

The idea of this plot is similar to gplot.hexbin. The hexagons are better suited to reflect the density of points in a plane than the squares used here. Nevertheless squares are, contrary to hexagons, invariant to reflexions at the x- and y-axis and therefore suited for scatter plot matrices and also for plotting factors.

Value

Maximum number of counts per Pixel found.

Note

When you get the error message "Zmax too small! Densiest aereas are out of range!" you must run the function again without specifying zmax. The value returned gives you the minimum value allowed for zmax.

Author(s)

Andreas Ruckstuhl, Rene Locher

See Also

ilagplot, ipairs, Image

Examples

x <-  rnorm(10000)
y <- atan(rnorm(10000, 0))
iplot(x, y)
iplot(x, pixs = 2)

oma <- c(5,5,0,0)
iplot(iris[,1:2], pixs = 4, oma = oma)
iplot(iris[,"Petal.Width"], iris[,"Species"], pixs = 4, oma = oma)
iplot(x = iris[,"Species"], y = iris[,"Petal.Width"], pixs = 10,border = TRUE,
        xlab = "Species",
        ylab = "Petal Width",
        main = "Iris Species and Petal Width", oma = oma)

iplot(iris$Species, iris$Petal.Width,pixs = 3, minL.axis = 10,
      oma = c(3,6,0,0), mgp = c(4, 1, 0),
      cex.axis = 2, cex.lab = 2, cex.main = 2, main = "Larger fonts")

Plots Legend for Color Ramp

Description

Plots legend for color ramp.

Usage

iplotLegend(colramp, ncol = NULL,
            cex.axis = par("cex.axis"), border = FALSE,
            mar = c(0, 0, 0, 3), las = 1, ...)

Arguments

Details

A color ramp of ncol+1 individual colors is plotted, starting with the background color (= color 0). When ncol = NULL a color ramp of 101 individual colors is plotted. The color at the bottom is labeled by '0', the color at the top by 'max'.

Author(s)

Rene Locher

Examples

## Not run: 
par(oma = rep(2, 4))
layout(matrix(c(1, 2), ncol = 2),
      width = c(lcm(par("csi")*5*2.54), 1),
      heights = 1)
## End(Not run)
iplotLegend(IDPcolorRamp, ncol = 15)

Plot Regression Terms for Huge Datasets

Description

Plots regression terms against their predictors, optionally with standard errors and partial residuals in a density plot.

Usage

itermplot(model, data = NULL, envir = environment(formula(model)),
          partial.resid = FALSE, scale=0, pixs = 1,
          zmax=NULL, ztransf = function(x) {x}, colramp = IDPcolorRamp,
          terms = NULL, se = FALSE,
          xlabs = NULL, ylabs = NULL, main = NULL,
          col.term = "black", lwd.term = 2,
          col.se = "gray", lty.se = 2, lwd.se = 1,
          col.smth = "darkred", lty.smth = 2,
          lwd.smth = 2, span.smth = 2/3,
          ask = interactive() && nb.fig < n.tms &&
                .Device != "postscript",
          use.factor.levels = TRUE, smooth = NULL, ...)

Arguments

Details

itermplot is a modified version of termplot of R V2.3.1. Partial residuals are displayed here as a density plot and is therfore especially suited for models of huge datasets. The model object must have a predict method that accepts type=terms, eg glm in the base package, coxph and survreg in the survival package.

For the partial.resid=TRUE option it must have a residuals method that accepts type="partial", which lm, glm and gam do.

The data argument should rarely be needed, but in some cases termplot may be unable to reconstruct the original data frame. Using na.action=na.exclude makes these problems less likely.

Nothing sensible happens for interaction terms.

Value

Maximum number of counts per pixel found.

Author(s)

Rene Locher

See Also

termplot.

Examples

r.lm <- lm(Sepal.Length~Sepal.Width+Petal.Length+Petal.Width+Species,
           data=iris)
par(mfrow=c(2,2),pty="s")
itermplot(r.lm, se = TRUE, partial.res=TRUE, lwd.term = 3,
          lwd.se = 2, pixs = 2)

if (require(SwissAir)) {
  data(AirQual)
  r.lm <- lm(log(ad.O3)~log(ad.NOx)+ad.T+ad.Td+ad.WS, data=AirQual)
  par(mfrow=c(2,2),pty="s")
  itermplot(r.lm, se = TRUE, partial.resid=TRUE, smooth=ipanel.smooth,
            lwd.smth = 3, pixs = 1, ask=FALSE)
} else print("Package SwissAir is not available")

Auxilary for Controlling the Appearance of the Legend of a Rose Plot

Description

Defines the appearance of the legend of a Rose Plot.

Usage

key.control(plot = TRUE, lab = NULL, title = NULL, between = 0)

Arguments

Value

Returns the arguments conveniently packaged up in a list to supply the arguments for the legend in plot.rose

Author(s)

Rene Locher

See Also

plot

Examples

key.control()

Plot Very Long Regular Time Series

Description

Plot one or more regular time series in multiple figures on one or more pages.

Usage

longtsPlot(y1, y2 = NULL,
           names1 = NULL, names2 = NULL,
           startP = start(y1)[1], upf = 400, fpp = 4, overlap = 20,
           x.at = NULL, x.ann = NULL, x.tick = NULL,
           y1.at = NULL, y1.ann = NULL, y1.tick = NULL,
           y2.at = NULL, y2.ann = NULL, y2.tick = NULL,
           nx.ann = 10, ny.ann = 3, cex.ann = par("cex.axis"),
           xlab = "", y1lab = "", y2lab = "", las = 0,
           col.y1 = "black", col.y2 = col.y1,
           cex.lab = par("cex.lab"),
           y1lim = range(y1, na.rm = TRUE, finite=TRUE),
           y2lim = range(y2, na.rm = TRUE, finite=TRUE),
           lty1 = 1, lty2 = 2, lwd1 = 1, lwd2 = lwd1,
           col1 = NULL, col2 = NULL,
           leg = TRUE, y1nam.leg = NULL, y2nam.leg = NULL,
           ncol.leg = NULL, cex.leg = par("cex"),
           h1 = NULL, h2 = NULL, col.h1 = "gray70", col.h2 = "gray70",
           main = NULL, cex.main = par("cex.main"),
           automain = is.null(main),
           mgp = c(2, 0.7, 0), mar = c(2,3,1,3)+.2,
           oma = if (automain|!is.null(main))
                             c(0,0,2,0) else par("oma"),
           xpd = par("xpd"), cex = par("cex"),
           type1 = "s", type2 = type1,
           pch1 = 46, pch2 = pch1, cex.pt1 = 2, cex.pt2 = cex.pt1,
           slide = FALSE, each.fig = 1,
           filename = NULL, extension = NULL, filetype = NULL, ...)

Arguments

Details

For longer time-series, it is sometimes important to spread several time-series plots over several subplots or even over several pages with several subplots in each. Moreover, these series have often different ranges, frequencies and start times. There is sometimes also the need of a more flexible annotation of axes than plot.ts provides. longtsPlot provides the user with all these features for one or two matrices or regular time series (time series matrices).

Side Effects

One or more pages of time series plots are drawn on the current graphic device and, optionally, saved in one or more files.

Author(s)

Rene Locher

Examples

## sunspots, y-axis only on the left
data(sunspots)
longtsPlot(sunspots,upf=ceiling((end(sunspots)-start(sunspots))[1]/5))

## air quality (left axis) and meteo data (right axis)
## use xpd=TRUE for time series with rare but large values
if (require(SwissAir)) {
  data(AirQual)
  st <- 6.5*30*48
  x.at <- seq(st,nrow(AirQual),48)
  longtsPlot(y1=AirQual[,c("ad.O3","ad.NOx")], y2 = AirQual$ad.T,
             names1=c("O3","NOx"),names2 = "Temp",
             startP = st, upf=7*48,
             x.at = x.at, x.ann = substr(AirQual$start,1,6)[x.at],
             x.tick = seq(st,nrow(AirQual),12),
             y1.at = c(0,100), y1.tick = seq(0,150,50),
             y2.at = c(10,30), y2.tick = seq(10,30,10),
             y1lab="[ppb]", y2lab="[C]",
             y1lim = c(0,100), y2lim = c(10,30), xpd=TRUE,
             col2 = "red", type1 = "l")
}

## Two time series with different frequencies and start times
## on the same figures
set.seed(13)
len <- 4*6*400
x <- sin((1:len)/200*pi)
d <- sin(cumsum(1+ rpois(len, lambda= 2.5)))

y1 <- ts(10*x,start=0,frequency=6)+d*rnorm(len)
y2 <- ts(100*x,start=100,frequency=13)+10*rnorm(len)
longtsPlot(y1,y2)

## plot your own legend
longtsPlot(sunspots, upf = ceiling((end(sunspots)-start(sunspots))[1]/5),
           fpp = 1, leg = FALSE)
legend(1750, 260, legend = "Monthly Sunspot Numbers", col = "blue", lwd = 1,
       bty = "n")

Sets NAs in Logical Objects to FALSE

Description

Sets NAs in logical vectors and matrices to FALSE. This is especially useful for conditional selections of data when the variables the condition is based on contain NAs.

Usage

ok(x)

Arguments

Value

Logical vector or matrix, containing no NAs.

Author(s)

Rene Locher

Examples

ok(c(FALSE,TRUE,NA,TRUE))

Finding Peaks in Raw Data

Description

Returns position, signal height and approximate width at half maximum peak height.

Usage

peaks(x, y = NULL, minPH, minPW, thr, stepF = 0.49)

Arguments

Details

The function is especially useful for signals in which both very broad and very narrow peaks are of interest. The peaks may lie very close to each other or might even be superpositioned on top of each other, e.g. peaks on broader shoulders. The algorithm is also very useful when the resolution of the signal is poor and the noise is small.

The function is looking for peaks without any preceding baseline substraction or smoothing, which could distort the spectrum.

The selection criteria minPH and minPW and the values for the calculated peak widths are only approximate.

Value

dataframe consisting of

Note

In the function, the main selection criterium for the peaks is the height of the peaks, the second optional criterium is the width of the peaks.

Author(s)

Rene Locher

Examples

n <- 200
freq <- 1:n
theory <- sin(freq/n*4*pi)*cos(freq/n*3*pi)
spec <- theory + 0.1*rnorm(n)

plot(spec,type="b")
lines(theory,lwd=2)

pts <- peaks(spec, minPH=0.7)
points(pts,col="red",cex=1.2, pch=20)

## peaks after smoothing the spectrum
spec.sm <- loess.smooth(freq, spec, span=0.2,
                        degree = 2, evaluation = 100)
lines(spec.sm$x, spec.sm$y, col="steelblue", lwd=2)
pts <- peaks(spec.sm, minPH=0.4)
points(pts,col="green",cex=1.2,pch=20)

## Analyses of Mass Spectrum between 12000 and 100'000
## without smoothing, without baseline substraction
data(MS)
MS1 <- log10(MS[MS$mz>12000&MS$mz<1e5,])

P <- peaks(MS1, minPH=0.02, minPW=0.001)
plot(MS1, type="l", xlab="log10(mz)", ylab="log10(I)")
points(P,col="blue",cex=1.6)

Plot Method for Class "rose" (Grid Graphics Function)

Description

Describes plot method for class "rose"

Usage

## S4 method for signature 'rose,missing'
plot(
          x,
          transf = function(x) sqrt(x),
          subset.col = NULL,
          warn = TRUE,
          general = general.control(),
          grid = grid.control(),
          title = title.control(),
          key = key.control())

Arguments

Details

This function appeals especially to environmental specialists who often have response variables, which depend from cyclic variables like the direction of wind, the hour of the day, the month etc. All these variables are displayed usually clockwise, starting with 0 in the north (12 o' clock). We call this kind of coordinates 'clock coordinates', to distinct them from the polar coordinates as used in mathematical context. The rose object is displayed as the time on a clock, measuring the angle defined in slot cyclVar in the clockwise direction from the north.
The eye takes the area of a graphical object as a measure of its size. This is why the default transformation of x@rho is chosen to be the square root. For equal distance roses use the transformation function function(x) x.
All labels, titles and line sizes are defined in multiples of cex.
This graphic function is based on package grid: Viewport vp.rose which was used to draw the rose and viewport vp.key which was used to draw the key may be addressed by pushviewport() after having drawn the figure.

Value

No value returned

Note

The function is designed to use the area on the active viewport in an optimal way, but the plot is not scalable after having been drawn.
Furthermore this function is still experimental so that some features may change in future versions.
You will find another nice example for this plot method in AirQual

Author(s)

Rene Locher

See Also

rose, rose-class

Examples

hour <- rep(0:23,100)
WD <- c(rnorm(24*90, mean=sample(c(190,220,50),24*90,
                                          replace = TRUE),sd=10),
                           rnorm(24*10, mean=360, sd=180))%%360
dat <- data.frame(A = (2*cos((hour+6)/6*pi)+
                       2*cos((WD+60)/180*pi)+rnorm(24*100,4))^2,
                  B = (2*cos((hour+4)/6*pi)+rnorm(24*100,1,8))^2)
dat$B[dat$B>1000] <- 1000

## two different response variables, scalar summary function
mean.dayrose <- rose(dat[,c("A","B")],
                cyclVar = hour,
                n.cyclVar = 24,
                circle = 24,
                FUN=mean, na.rm=TRUE)

## one response variable, vector summary function
quant.windrose <- rose(dat$A,
                       cyclVar = WD,
                       n.cyclVar = 16, circle = 360,
                       FUN=quantile, na.rm=TRUE)

## one response variable, second (non cyclic) explanatory variable,
## scalar summary function
windrose <- rose(dat[,c("A")],
                 cyclVar = WD,
                 n.cyclVar=8,
                 circle = 360,
                 cut = dat$B,
                 breaks = c(0,10,100,1000),
                 include.lowest = TRUE, dig.lab = 4,
                 FUN = function(x) sum(!is.na(x)))

grid::grid.newpage()
plot(mean.dayrose,
     general = general.control(
       mar = rep(1,4),
       stacked = FALSE,
       lwd = 3,
       lty = c(1:2)),
     grid = grid.control(
       circ.n = 2,
       circ.sub.n = 2,
       circ.lwd = 2,
       circ.sub.col = "black",
       ray.n = 12,
       cyclVar.lab = seq(0,by=2,to=22)),
     title = title.control(text = "unstacked dayrose"),
     key = key.control(title = "Mean",
                       between = 0))

grid::grid.newpage()
plot(quant.windrose)

grid::grid.newpage()
plot(windrose,
     general = general.control(
       stacked = TRUE,
       lwd = 3),
     grid = grid.control(
       circ.n = 2,
       circ.sub.n = 2),
     title = title.control(
       text = "Stacked windrose:\nCounts of A-Values"),
     key = key.control(title = "Value of B"))


if (require(SwissAir)){
  data(AirQual)
  dat <-
    data.frame(month =as.numeric(substr(AirQual$start,4,5)),
               hour = as.numeric(substr(AirQual$start,12,13)),
               WD = AirQual$ad.WD,
               NOx = AirQual$ad.NOx,
               ## NO2 = AirQual$ad.NOx-AirQual$ad.NO,
               ## NO  = AirQual$ad.NO,
               O3  = AirQual$ad.O3,
               Ox  = AirQual$ad.O3+AirQual$ad.NOx-AirQual$ad.NO)


  ## Windrose
  windrose <- rose(dat$WD,
                   cyclVar = dat$WD, n.cyclVar = 32, circle = 360,
                   FUN = function(x) sum(!is.na(x)))

  grid::grid.newpage()
  plot(windrose,
       general =
       general.control(lwd = 2),
       grid =
       grid.control(circ.n = 2,
                    circ.sub.n = 2))

  ## median of concentrations as a function of daytime
  ## from May to September
  med.dayrose <- rose(dat[,c("NOx","O3","Ox")],
                      subset= dat$month>4 & dat$month<10,
                      cyclVar=dat$hour, n.cyclVar=24, circle=24,
                      FUN=median, na.rm=TRUE)

  ## line type version of rose
  grid::grid.newpage()
  plot(med.dayrose,
       general = general.control(lwd=2, type="l"),
       grid =
       grid.control(ray.n = 12,
                    circ.n =2,
                    circ.sub.n = 2,
                    cyclVar.lab = seq(0,by=2,to=22)),
       title = title.control(text =
         "Day Rose of Medians\nduring summer"))

  ## quantiles of concentrations as a function of daytime
  ## from May to September
  quant.dayrose <- rose(dat$NOx,
                        subset= dat$month>4 & dat$month<10,
                        cyclVar=dat$hour, n.cyclVar=24, circle=24,
                        FUN=quantile, na.rm=TRUE)
  grid::grid.newpage()
  plot(quant.dayrose,
       general =
       general.control(mar = c(0.3, 0.3, 0.3, 2),
                       lwd = 2),
       grid =
       grid.control(ray.n = 12,
                    cyclVar.lab = seq(0,by=2,to=22)),
       title = title.control(text = "Concentration of NOx [ppb]\nduring summer"),
       key = key.control(title = "Quantiles"))
} else print("Package SwissAir is not available")

Convenient xyplot with Differently Colored Margin and Plot Region

Description

Convenient xyplot with Colored Background. Background of margin may be chosen independently from background in plot region.

Usage

poster.plot(x, y = NULL, type = "p",
            col = col.fg, col.axis = col.fg, col.lab = col.fg,
            col.fg = "blue", col.bg = "lavender", col.box = "cornsilk",
            xlim = NULL, ylim = NULL, xlab = "", ylab = "",
            main = "", cex = 1.2, axes = TRUE, ...)

Arguments

Details

Side effect: par options will remain changed so that other graphic elements can be added comfortably.

Author(s)

Andreas Ruckstuhl, refined by Rene Locher

Examples

poster.plot(iris[,1],iris[,2],
            xlab="Sepal.Length", ylab="Sepal.Width")

poster.plot(iris[,1], col="red", col.box="grey95", ylab="Sepal.Length")

## plotting lines
n <- 200
freq <- 1:n
y <- sin(freq/n*4*pi)*cos(freq/n*3*pi) + 0.1*rnorm(n)
poster.plot(y,col.fg="grey30",type="l")

Robust Fitting of Baselines

Description

Robust fitting of local regression models for estimating a baseline or a background signal

Usage

rfbaseline(x, y, span = 2/3, NoXP = NULL,
           maxit = c(2, 2), b = 3.5, weight = NULL,
           Scale = function(r) median(abs(r))/0.6745,
           delta = NULL, SORT = TRUE, DOT = FALSE, init = NULL)

Arguments

Value

List containing components

Author(s)

Andreas Ruckstuhl

References

Ruckstuhl, Andreas F., Matthew P. Jacobson, Robert W. Field and James A. Dodd (2001); Baseline Subtraction Using Robust Local Regression Estimation; Journal of Quantitative Spectroscopy and Radiative Transfer 68: 179 – 193

Ruckstuhl, Andreas F., et al.; Estimation of background concentrations of atmospheric trace gases using robust local regression; to be published

See Also

See Also as loess and lowess

Examples

data(MS)
MS1 <- log10(MS[MS$mz>12000&MS$mz<1e5,])

MS1.rfb2 <- rfbaseline(x=MS1$mz, y=MS1$I, NoXP=2200, maxit=c(5,0))
plot(x=MS1$mz, y=MS1$I, type="l",
     xlab="log(mass/charge)", ylab="log(intensity)")
lines(MS1.rfb2$x, MS1.rfb2$fit, col="orange", lwd=3)

MS1.rfb3 <- rfbaseline(x=MS1$mz, y=MS1$I, NoXP=1100, maxit=c(5,0),
                       DOT=TRUE, Scale=function(x) mad(x, center=0))
plot(x=MS1$mz, y=MS1$I, type="l",
     xlab="log(mass/charge)", ylab="log(intensity)")
lines(MS1.rfb3$x, MS1.rfb3$fit, col="orange", lwd=3)

## 'delta=0' needs much more computer time
## Not run: 
MS1.rfb4 <- rfbaseline(x=MS1$mz, y=MS1$I, NoXP=2200,
                       delta=0, maxit=c(5,0))
plot(x=MS1$mz, y=MS1$I,ty="l",
     xlab="log(mass/charge)", ylab="log(intensity)")
lines(MS1.rfb4$x, MS1.rfb4$fit, col="orange", lwd=3)

## End(Not run)

Estimation of the Scale Parameter

Description

Estimation of the scale parameter based on data smaller than its first mode. Mainly used in rfbaseline.

Usage

rfbaselineScale(r)

Arguments

Value

Estimated scale.

Author(s)

Andreas Ruckstuhl

See Also

See also mad

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

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

See Also

rose-class, plot.rose, cart2clock, clock2cart

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

rose-class

Description

Summary statistics of cyclic data.

Objects from the Class

Objects can be created by calls of the form rose(x, cyclVar = NULL, circle = NULL, n.cyclVar = 8, cut = NULL, breaks = NULL, labels = NULL, dig.lab = 2, include.lowest = FALSE, subset = NULL, na.warning = TRUE, FUN = mean, ...)

Slots

rho:

Object of class matrix. rho[i,] contains the summary values of all data within the interval defined by the cyclic Variable cyclVar[i]. Column and row names are mandatory. The different columns of rho correspond to different responses when the summary statistics is a scalar or to the different elements of a vector summary of one response or to the different subsets of the second grouping element.

cyclVar:

Object of class numeric containing the center of the interval of the cyclic variable. The values are sorted by increasing values, are unique and cannot contain NA values.

circle:

Scalar of class numeric, defining the full circle.

Author(s)

Rene Locher

See Also

For the details of how to create and plot a rose object see rose and plot.rose.

Displays vectors of colors

Description

Displays colors produced by a color vector col and labels them by the corresponding number of the element of col.

Usage

showColors(col = IDPcolorRamp(20),
           ntm = min(length(col), 20),
           border = TRUE, mar = rep(0,4))

Arguments

Author(s)

Rene Locher

See Also

IDPcolorRamp, ColorBrewer

Examples

showColors(IDPcolorRamp(5))
showColors(IDPcolorRamp(200),border=FALSE)
showColors(IDPcolorRamp(200),border=FALSE,ntm=5)

showColors(IDPcolorRamp(4,
                        colInt = data.frame(
                            h = c(0.47, 0.28, 0.22, 0.2, 0.00),
                            s = c(0.3, 0.55, 0.75, 0.75, 0.75),
                            v = c(1, 1, 1, 1, 1)),
                        fr     = c(0.2,0.2,0.2))
           ,border=FALSE)

showColors(IDPcolorRamp(200,
                        t(col2hsv(c("darkviolet","blue",
                                    "green","yellow","red"))),
                        fr=rep(0.25,3)),border=FALSE)

Auxilary for Controlling the Title of a Rose Plot

Description

Auxilary for controlling the title of a rose plot.

Usage

title.control(text = NULL, cex = 1.5,
              between = if (is.null(text)) 0 else 1)

Arguments

Value

Returns the arguments conveniently packaged up in a list to supply the arguments for the title of plot.rose.

Author(s)

Rene Locher

See Also

plot.rose, general.control

Examples

title.control()

Zooming in and out in a 2d-Plot

Description

Function to zoom in and out by mouse click in a 2D-plot.

Usage

zoom(fun = plot, zoom.col = "red", delay = 3, ...)

Arguments

Details

When the clicked points lay within the plot region range, the points define the new plotting limits. When the clicked points lay in the margin, the plotting limits will be moved into the corresponding direction by 1/3 out of the actual range. There is no special sequential order for the zooming points required. The zooming function is stopped by right clicking and choosing the menu item "stop".

Value

No value returned.

Author(s)

Rene Locher

Examples

i <- 1:100
y <- i*sin(i*(pi/16))
y <- c(rev(y),y)

## Not run:
zoom(fun=plot, zoom.col="red", x=1:200, y=y, type="l", xlab="index")
## End( Not run)