Package 'smvgraph' reference manual

Title:	Visualization and Clustering of Data in a Shiny App
Description:	Various visualisations of univariate and multivariate graphs (e.g. mosaic diagram, scatterplot matrix, Andrews curves, parallel coordinate diagram, radar diagram and Chernoff plots) as well as clustering methods (e.g. k-means, agglomerative, EM clustering and DBSCAN) are implemented as a Shiny app. The app allows interactive changes, e.g. of the order of variables. It is intended for use in teaching.
Authors:	Sigbert Klinke [aut, cre]
Maintainer:	Sigbert Klinke <[email protected]>
License:	GPL-3
Version:	0.2.0
Built:	2025-02-02 03:26:05 UTC
Source:	https://github.com/sigbertklinke/smvgraph

andrews

Description

Andrews curves for visualization of multidimensional data. step determines the number of line segments for each curve. If ymax==NA then the maximum y coordinate will be determined from the curves. Note that for type==3 the x range is $[0, 4*pi]$ otherwise $[-pi, pi]$ . Observations containing NA, Nan, -Inf, or +Inf will be deleted before plotting

Usage

andrews(x, type = 1, step = 100, ..., normalize = 1, ymax = NA)
andrews(x, type = 1, step = 100, ..., normalize = 1, ymax = NA)

Arguments

`x`	data frame or matrix
`type`	type of curve (default: `1`) 1: $f(t)=x1/(2^0.5)+x2sin(t)+x3cos(t)+x4sin(2t)+x5cos(2t)+...$ 2: $f(t)=x1sin(t)+x2cos(t)+x3sin(2t)+x4cos(2t)+...$ 3: $f(t)=x1cos(t)+x2cos((2t)^0.5)+x3cos((3t)^0.5)+...$ 4: $f(t)=1/(2^0.5)(x1+x2(sin(t)+cos(t))+x3(sin(t)-cos(t))+x4(sin(2t)+cos(2t))+x5(sin(2t)-cos(2t))+...)$
`step`	smoothness of curves
`...`	further parameters given to graphics::plot and graphics::lines
`normalize`	integer: normalization method (default: `1`) 0: no rescaling 1: $(x-min(x))/(max(x)-min(x))$ 2: $(x-mean(x))/sd(x)$
`ymax`	numeric: maximum of y coordinate (default: `NA`)

Value

nothing

References

Andrews, D. F. (1972) Plots of High-Dimensional Data. Biometrics, vol. 28, no. 1, pp. 125-136.
Khattree, R., Naik, D. N. (2002) Andrews Plots for Multivariate Data: Some New Suggestions and Applications. Journal of Statistical Planning and Inference, vol. 100, no. 2, pp. 411-425.

Examples

andrews(iris[,-5], col=as.factor(iris[,5]))
andrews(iris[,-5], type=4, col=as.factor(iris[,5]), ymax=2)
andrews(iris[,-5], col=as.factor(iris[,5]))
andrews(iris[,-5], type=4, col=as.factor(iris[,5]), ymax=2)

as_param

Description

Create a parameter list or a function call. For a function call fun must be explicitly given.

Usage

as_param(..., fun = NULL)

txt(x)
as_param(..., fun = NULL)

txt(x)

Arguments

`...`	list of named and unnamed parameters
`fun`	character:
`x`	character: replaces `"x"` by `"'x'"`

Value

a character as parameter list of function call

Examples

as_param(letters[1:5])
as_param(txt(letters[1:5]))
as_param(a=txt("a"))
as_param(txt(letters[1:5]), fun="c")
as_param(letters[1:5])
as_param(txt(letters[1:5]))
as_param(a=txt("a"))
as_param(txt(letters[1:5]), fun="c")

availablePlots

Description

Returns a data frame with columns about the available plots in smvgraph:

module: the internal name used. If you want to call the Shiny app then you might need this.
label: the label used in the Shiny app
help: the R help topic for the plot
packages: packages which are required to make the plot
code: if code block exists, should always be TRUE
ui: if plot specific interactive UI elements exists
condition: the condition when a plot is offered in the Shiny app to the user

Usage

availablePlots()
availablePlots()

Details

To understand condition:

nrow(analysis): the number of variables in "Analysis" field
nrow(group): the number of variables in "Grouping by" field
xxx$unique: the number of unique values in a variable, for other elements then unique see the "Variable" panel of the Shiny app

13 was choosen because twelve has the largest number of divisors below 20 and 43 was choosen because forty-two is the answer of the ultimate question ;)

Value

a data frame with information about all available plots

Examples

availablePlots()
availablePlots()

bagplot

Description

A non-ggplot2 bagplot based on mrfDepth::bagplot.

Usage

bagplot2(
  x,
  y = NULL,
  colorbag = NULL,
  colorloop = NULL,
  colorchull = NULL,
  databag = TRUE,
  dataloop = TRUE,
  plot.fence = FALSE,
  type = "hdepth",
  sizesubset = 500,
  extra.directions = FALSE,
  options = NULL,
  ...
)
bagplot2(
  x,
  y = NULL,
  colorbag = NULL,
  colorloop = NULL,
  colorchull = NULL,
  databag = TRUE,
  dataloop = TRUE,
  plot.fence = FALSE,
  type = "hdepth",
  sizesubset = 500,
  extra.directions = FALSE,
  options = NULL,
  ...
)

Arguments

`x`, `y`	the x and y arguments provide the x and y coordinates for the bagplot. Any reasonable way of defining the coordinates is acceptable. See the function `xy.coords` for details. If supplied separately, they must be of the same length.
`colorbag`	The color of the bag (which contains the 50% observations with largest depth).
`colorloop`	The color of the loop (which contains the regular observations).
`colorchull`	When the bagplot is based on halfspace depth, the depth region with maximal depth is plotted. This argument controls its color.
`databag`	Logical indicating whether data points inside the bag need to be plotted. Defaults to `TRUE`.
`dataloop`	Logical indicating whether data points inside the fence need to be plotted. Defaults to `TRUE`.
`plot.fence`	Logical indicating whether the fence should be plotted. Defaults to `FALSE`.
`type`	Determines the depth function used to construct the bagplot: `"hdepth"` for halfspace depth, `"projdepth"` for projection depth and `"sprojdepth"` for skewness-adjusted projection depth. Defaults to `"hdepth"`.
`sizesubset`	When computing the bagplot based on halfspace depth, the size of the subset used to perform the main computations. See Details for more information. Defaults to $500$ .
`extra.directions`	Logical indicating whether additional directions should be considered in the computation of the fence for the bagplot based on projection depth or skewness-adjusted projection depth. If set to `TRUE` an additional 250 equispaced directions are added to the directions defined by the points in `x` themselves and the center. If `FALSE` only directions determined by the points in `x` are considered. Defaults to `FALSE`.
`options`	A list of options to pass to the `projdepth` or `sprojdepth` function. In addition the following option may be specified: `max.iter` The maximum number of iterations in the bisection algorithm used to compute the depth contour corresponding to the cutoff. See `depthContour` for more information. Defaults to $100$ .
`...`	further parameters given to `plot`

Details

The bagplot has been proposed by Rousseeuw et al. (1999) as a generalisation of the boxplot to bivariate data. It is constructed based on halfspace depth and as such is invariant under affine transformations. Similar graphical representations can be obtained by means of other depth functions, as illustrated in Hubert and Van der Veeken (2008) and in Hubert et al. (2015). See mrfDepth::compBagplot for more details.

The deepest point is indicated with a "*" sign, the outlying observations with red points.

Value

Invisibly the result of the call to mrfDepth::compBagplot

References

Rousseeuw P.J., Ruts I., Tukey, J.W. (1999). The bagplot: a bivariate boxplot. The American Statistician, 53, 382–387.

Hubert M., Van der Veeken S. (2008). Outlier detection for skewed data. Journal of Chemometrics, 22, 235–246.

Hubert M., Rousseeuw P.J., Segaert, P. (2015). Rejoinder to 'Multivariate functional outlier detection'. Statistical Methods & Applications, 24, 269–277.

Examples

bagplot2(iris$Sepal.Length, iris$Sepal.Width)
bagplot2(iris[,1:2])
bagplot2(iris[,3:4], title="Bagplot with Tukey depth", xlab="Petal.Length", ylab="Petal.Width") 
#
library("mrfDepth")
data("bloodfat")
result <- compBagplot(bloodfat)
bagplot(result, colorbag = rgb(0.2,0.2,0.2), colorloop = "green")
bagplot2(iris$Sepal.Length, iris$Sepal.Width)
bagplot2(iris[,1:2])
bagplot2(iris[,3:4], title="Bagplot with Tukey depth", xlab="Petal.Length", ylab="Petal.Width") 
#
library("mrfDepth")
data("bloodfat")
result <- compBagplot(bloodfat)
bagplot(result, colorbag = rgb(0.2,0.2,0.2), colorloop = "green")

binData

Description

Bins each variable in data in bins Bins. It can return a data frame (out="data.frame"), a table with the counts (out="table"), or a table converted to a data frame with an additional variable Freq. The values can be either the bin mids (val="mids") or the bin numbers (val="interval"). If possible all variables contain an attribute breaks with breaks used.

Usage

binData(
  data,
  bins,
  out = c("data.frame", "table", "binned"),
  val = c("mid", "interval"),
  pretty = TRUE,
  numeric = TRUE
)
binData(
  data,
  bins,
  out = c("data.frame", "table", "binned"),
  val = c("mid", "interval"),
  pretty = TRUE,
  numeric = TRUE
)

Arguments

`data`	object: a data.frame or object that can be converted to a data frame with variables to bin
`bins`	integer: number of bins, will be recycled if necessary
`out`	character: output type, either `"data.frame"`, `"table"`, or `"binned"`
`val`	character: values for outer, eiter `"mids"` (interval centers), or `⁠"interval⁠` (interval number)
`pretty`	logical: should be base::pretty used or minimum and maximum (default: `TRUE`)
`numeric`	logical: return output a `factor` or as `numeric` (default: `TRUE`)

Value

a data frame or table with the results

Examples

df <- data.frame(x=runif(25), y=runif(25))
binData(df, 5, 'data.frame')
binData(df, 5, 'table')
binData(df, 5, 'binned')
df <- data.frame(x=runif(25), y=runif(25))
binData(df, 5, 'data.frame')
binData(df, 5, 'table')
binData(df, 5, 'binned')

character_data

Description

Converts a matrix or data frame into a character vector, matrix or data frame. If na.action is a character then all NAs are replaced by na.action (default: na.action="NA"). If na.action is a function then the function will be applied to the result.

Usage

character_data(
  x,
  select = NULL,
  out = c("data.frame", "matrix", "vector"),
  na.action = "NA",
  ...,
  title = NULL
)
character_data(
  x,
  select = NULL,
  out = c("data.frame", "matrix", "vector"),
  na.action = "NA",
  ...,
  title = NULL
)

Arguments

`x`	vector, matrix or data frame
`select`	vector: indicating columns to select (default: `NULL`)
`out`	output as `data.frame` (default), `matrix`, or `vector`
`na.action`	function or character: indicates what should happen when the data contain NAs
`...`	unused
`title`	character: title attribute (default `NULL`)

Value

the desired R object

Examples

character_data(iris)
character_data(iris, out="matrix")
character_data(iris, out="vector")
character_data(iris)
character_data(iris, out="matrix")
character_data(iris, out="vector")

checkPackages, installPackages

Description

Checks if a package is installed without loading it. Returns a logical vector with TRUE or FALSE for each package checked.

Usage

checkPackages(
  ...,
  plotmodule = NULL,
  add = c("tools", "devtools", "formatR", "highlight", "shiny", "shinydashboard",
    "shinydashboardPlus", "shinyWidgets", "DT", "sortable", "base64enc"),
  error = FALSE
)

installPackages(
  plotmodule = NULL,
  add = c("tools", "devtools", "formatR", "highlight", "shiny", "shinydashboard",
    "shinydashboardPlus", "shinyWidgets", "DT", "sortable", "base64enc")
)
checkPackages(
  ...,
  plotmodule = NULL,
  add = c("tools", "devtools", "formatR", "highlight", "shiny", "shinydashboard",
    "shinydashboardPlus", "shinyWidgets", "DT", "sortable", "base64enc"),
  error = FALSE
)

installPackages(
  plotmodule = NULL,
  add = c("tools", "devtools", "formatR", "highlight", "shiny", "shinydashboard",
    "shinydashboardPlus", "shinyWidgets", "DT", "sortable", "base64enc")
)

Arguments

`...`	character: name(s) of package
`plotmodule`	character: name(s) of plot modules to check for packages
`add`	character: names of default packages to check (default: `c("highlight", "formatR", "shiny", "shinydashboard", "shinydashboardPlus", "DT")`)
`error`	logical: should a error thrown if one or more package are missing? (default: `FALSE`)

Value

TRUE if successful otherweise an error will be thrown

Examples

checkPackages("graphics", add=NULL)          # checks if 'graphics' is installed
if (interactive()) checkPackages("graphics") # checks if 'graphics', 'shiny', ... are installed
if (interactive()) installPackages()         # installs all packages to show ALL plots
checkPackages("graphics", add=NULL)          # checks if 'graphics' is installed
if (interactive()) checkPackages("graphics") # checks if 'graphics', 'shiny', ... are installed
if (interactive()) installPackages()         # installs all packages to show ALL plots

color_data

Description

Assigns a color to the data x based on the color palette colpal.

Usage

color_data(x, colpal = grDevices::hcl.colors, select = NULL, ..., title = NULL)
color_data(x, colpal = grDevices::hcl.colors, select = NULL, ..., title = NULL)

Arguments

`x`	vector, matrix, or data frame
`colpal`	color palette (default: grDevices::hcl.colors)
`select`	vector: indicating columns to select (default: `1`)
`...`	further parameters to factor_data
`title`	character: title attribute (default `NULL`)

Value

a color vector

Examples

color_data(iris)
color_data(iris$Species)
color_data(iris)
color_data(iris$Species)

color_hclust

Description

Determines colors for x based on stats::hclust. x is normalized according normalize.

Usage

color_hclust(
  x,
  normalize = 1,
  ncol = 2,
  colpal = grDevices::hcl.colors,
  dist = "euclidean",
  na.action = stats::na.pass,
  ...
)
color_hclust(
  x,
  normalize = 1,
  ncol = 2,
  colpal = grDevices::hcl.colors,
  dist = "euclidean",
  na.action = stats::na.pass,
  ...
)

Arguments

`x`	a numeric matrix, data frame or "dist" object.
`normalize`	integer: normalization method (default: `1`) 0: no rescaling 1: $(x-min(x))/(max(x)-min(x))$ 2: $(x-mean(x))/sd(x)$
`ncol`	integer: maximal number colors
`colpal`	color palette: a function which generates "ncol" colors with "colpal(ncol)" (default: grDevices::hcl.colors)
`dist`	the distance measure to be used. This must be one of "euclidean", "maximum", "manhattan", "canberra" or "binary"(default: `euclidean`)
`na.action`	a function which indicates what should happen when the data contain NAs (default: `na.pass`)
`...`	further parameters given to stats::hclust

Value

a color vector

Examples

color_hclust(iris[,-5], ncol=6)
color_hclust(iris[,-5], ncol=6)

convertTo

Description

Converts an input object (vector, matrix or data frame) to an output asccording to the format in out. Variable, Row and column names are set, if possible as well as a attribute title.

Usage

convertTo(x, coln, rown, title, out = c("data.frame", "matrix", "vector"))
convertTo(x, coln, rown, title, out = c("data.frame", "matrix", "vector"))

Arguments

`x`	vector, matrix, or data frame: input
`coln`	character: column names if possible
`rown`	character: row names if possible
`title`	character: for title attribute
`out`	character: either

Value

the desired output object

Examples

str(convertTo(pi, "Col1", "Row1", "Title", out='data.frame'))
str(convertTo(pi, "Col1", "Row1", "Title", out='matrix'))
str(convertTo(pi, "Col1", "Row1", "Title", out='vector'))
str(convertTo(pi, "Col1", "Row1", "Title", out='data.frame'))
str(convertTo(pi, "Col1", "Row1", "Title", out='matrix'))
str(convertTo(pi, "Col1", "Row1", "Title", out='vector'))

factor_data

Description

Creates a single group variable from the data x.

Usage

factor_data(
  x,
  select = NULL,
  out = c("data.frame", "matrix", "vector"),
  exclude = NULL,
  na.action = stats::na.pass,
  ...,
  title = NULL
)
factor_data(
  x,
  select = NULL,
  out = c("data.frame", "matrix", "vector"),
  exclude = NULL,
  na.action = stats::na.pass,
  ...,
  title = NULL
)

Arguments

`x`	vector, matrix, or data frame
`select`	vector: indicating columns to select (default: `NULL`)
`out`	output as `data.frame` (default), `matrix`, or `vector`
`exclude`	vector: values to be excluded when forming the set of levels (default: `NULL`)
`na.action`	a function which indicates what should happen when the data contain NAs (default: stats::na.pass)
`...`	further parameters to character_data
`title`	character: title attribute (default `NULL`)

Value

a one-column matrix with the merged groups

Examples

factor_data(iris$Species, out="vector")
factor_data(iris)
factor_data(iris$Species, out="vector")
factor_data(iris)

formatCommands

Description

formatCommands

Usage

formatCommands(cmds)
formatCommands(cmds)

Arguments

cmds

characater: R code

Value

HTML code for the splot app

Examples

formatCommands('print("Hello World!")')
formatCommands('print("Hello World!")')

getModules

Description

Returns a list of available module as a list.

Usage

getModules(pattern, path = getShinyOption("smvgraph.path"))
getModules(pattern, path = getShinyOption("smvgraph.path"))

Arguments

`pattern`	character: character string containing a regular expression, currently are used `plot_.R` and `color_.R`
`path`	character: containing the path where to search the modules

Value

a list with the modules

Examples

library("shiny")
getModules('plot_*.R')   # get plots
getModules('color_*.R')  # get colors
library("shiny")
getModules('plot_*.R')   # get plots
getModules('color_*.R')  # get colors

getval

Description

Returns val if length(val)>1. Otherwise it runs through args=list(...) until it finds an element with length(args[[i]])>0 and returns it. If everything fails NULL will be returned.

Usage

getval(val, ...)
getval(val, ...)

Arguments

`val`	current value
`...`	sequence of alternative values

Value

a value

Examples

getval(NULL, 0)
getval(1, 0)
getval(NULL, 0)
getval(1, 0)

getVariableInfo

Description

Returns a data frame with one row for each variable in data:

Usage

getVariableInfo(data, n = 47)
getVariableInfo(data, n = 47)

Arguments

`data`	data frame: input data set
`n`	integer: character length for `values` (default: `47`)

Details

class the base::class of the variable
missing the number of missing values
infinite the number of infinite values
unique the number of unique values
valid the number of unique valid values (see valid)
values the values with the decreasing frequency

Value

a data frame with information about the variables of the input data set

Examples

getVariableInfo(iris)
getVariableInfo(iris)

getVariableNames

Description

Extracts variable names from a data frame or matrix (column names).

Usage

getVariableNames(x, xvar = NULL, num = TRUE)
getVariableNames(x, xvar = NULL, num = TRUE)

Arguments

`x`	data frame/matrix: data set to analyse
`xvar`	character: variable names to analyse (default: `character(0)` = all variables)
`num`	logical: should numerical or non-numerical variable use (default: `TRUE`)

Value

character vector with variable names

Examples

getVariableNames(iris)
getVariableNames(iris, num=FALSE)
getVariableNames(normalize(iris, 0))
getVariableNames(normalize(iris, 0), num=FALSE)
getVariableNames(iris)
getVariableNames(iris, num=FALSE)
getVariableNames(normalize(iris, 0))
getVariableNames(normalize(iris, 0), num=FALSE)

jitter_min

Description

Add a small amount of noise to a numeric vector. The result is x + runif(n, -a, a) where n <- length(x) and a <- abs(factor*amount) argument. If amount==0 then amount is set to 1e-6 times the smallest non-zero distance between adjacent unique x values. In case of no non-zero distances amount is set to 1e-6*(1+min(abs(x))). Note that jitter_min delivers different results then base::jitter.

Usage

jitter_min(x, factor = 1, amount = 0)
jitter_min(x, factor = 1, amount = 0)

Arguments

`x`	numeric: vector to which jitter should be added
`factor`	numeric: multiplier for `amount` (default: `1`)
`amount`	numeric: amount for jittering (default: `0`)

Value

jittered data

Examples

jitter_min(runif(6))
jitter_min(rep(0, 7))
jitter_min(rep(10000, 5))
jitter_min(runif(6))
jitter_min(rep(0, 7))
jitter_min(rep(10000, 5))

loggit

Description

Stores in a temporary file the log messages including messages, warnings and errors.

Usage

loggit(log_lvl, log_msg)

read_logs()

set_logfile()
loggit(log_lvl, log_msg)

read_logs()

set_logfile()

Arguments

`log_lvl`	character: Level of log output. In actual practice, one of "DEBUG", "INFO", "WARN", and "ERROR" are common, but any string may be supplied
`log_msg`	character: Main log message

Value

Nothing.

Examples

if (interactive()) {
  set_logfile()      # create a temporary file for logging
  loggit("DEBUG", "Hello world")
  read_logs()        # get a data frame with the current messages.
}
if (interactive()) {
  set_logfile()      # create a temporary file for logging
  loggit("DEBUG", "Hello world")
  read_logs()        # get a data frame with the current messages.
}

normalize

Description

Extracts the numeric vectors from a data frame and normalizes each vector. Note: In case that a variable is constant for method==1 (minmax) the entries will be replaced by 0.5 and for method==2 (standardization) the entries will be replaced by 0.

Usage

normalize(x, method = 1)
normalize(x, method = 1)

Arguments

x

data.frame or matrix

method

integer: normalization method (default: 1)

0: no rescaling
1: $(x-min(x))/(max(x)-min(x))$
2: $(x-mean(x))/sd(x)$

Value

numeric matrix

Examples

normalize(iris, 2)
normalize(iris, 2)

numeric_data

Description

Converts a vector, matrix or data frame into a numeric vector, matrix or data frame.

Usage

numeric_data(
  x,
  select = NULL,
  out = c("data.frame", "matrix", "vector"),
  na.action = stats::na.pass,
  ...,
  title = NULL
)
numeric_data(
  x,
  select = NULL,
  out = c("data.frame", "matrix", "vector"),
  na.action = stats::na.pass,
  ...,
  title = NULL
)

Arguments

`x`	vector, matrix or data frame
`select`	vector: indicating columns to select (default: `NULL`)
`out`	output as `data.frame` (default), `matrix`, or `vector`
`na.action`	a function which indicates what should happen when the data contain NAs (default: stats::na.pass)
`...`	unused
`title`	character: title attribute (default `NULL`)

Value

the desired R object

Examples

numeric_data(iris)
numeric_data(iris, out="matrix")
numeric_data(iris, out="vector")
numeric_data(iris)
numeric_data(iris, out="matrix")
numeric_data(iris, out="vector")

order_andrews

Description

Returns a reording of the columns of x to visualize outliers or clusters better. If no colum names are given then V1, V2, ... will be used.

Usage

order_andrews(x, method = 1)
order_andrews(x, method = 1)

Arguments

x

data matrix

method

numeric: order method (default: 1)

1: interquartile range
2: $max(x-median(x))/IQR(x)$ (outlier)
3: fit to a Ward cluster solution with euclidean distance

Value

order of column vectors

Examples

order_andrews(iris)
order_andrews(iris)

order_parcoord

Description

Returns a reordering of the columns of x to visualize highly correlated variable pairs based on a cluster analysis of the correlation matrix. If no colum names are given then V1, V2, ... will be used.

Usage

order_parcoord(x, method = "spearman", ...)
order_parcoord(x, method = "spearman", ...)

Arguments

`x`	data matrix
`method`	numeric: order method (default: `"spearman"`)
`...`	further parameters given to stats::cor

Value

order of column vectors

Examples

order_parcoord(iris)
order_parcoord(iris)

pyramid

Description

pyramid

Usage

pyramid(
  tab,
  gap = 0,
  left = list(col = "red"),
  right = list(col = "blue"),
  ...
)
pyramid(
  tab,
  gap = 0,
  left = list(col = "red"),
  right = list(col = "blue"),
  ...
)

Arguments

`tab`	table: a table with two columns
`gap`	numeric(2): relative size of gap in `y`- and `x`-direction (default: `c(0,0)`)
`left`	list: parameters for the left polygons (default: `list(col="red")`)
`right`	list: parameters for the right polygons (default: `list(col="blue")`)
`...`	further parameters to use in graphics::plot.default

Value

a pyramid plot

Examples

data("Boston", package="MASS")
tab <- table(data.frame(Boston$rad, Boston$chas))
pyramid(tab, main="Absolute frequencies")
pyramid(tab, gap=c(0.2, 0.2))
rtab <- tab/sum(tab)
pyramid(rtab, gap=c(0.2, 0.2), main="Relative frequencies")
ctab <- proportions(tab, 2)
pyramid(ctab, gap=c(0.2, 0.2), main="Conditional frequencies on columns")
rtab <- proportions(tab, 1)
pyramid(rtab, gap=c(0.2, 0.2), main="Conditional frequencies on rows")
# zebraing 
pyramid(tab, gap=c(0.2, 0.2), 
        left=list(list(col="black"), list(col="white")), 
        right=list(list(col="blue"), list(col="green")))
data("Boston", package="MASS")
tab <- table(data.frame(Boston$rad, Boston$chas))
pyramid(tab, main="Absolute frequencies")
pyramid(tab, gap=c(0.2, 0.2))
rtab <- tab/sum(tab)
pyramid(rtab, gap=c(0.2, 0.2), main="Relative frequencies")
ctab <- proportions(tab, 2)
pyramid(ctab, gap=c(0.2, 0.2), main="Conditional frequencies on columns")
rtab <- proportions(tab, 1)
pyramid(rtab, gap=c(0.2, 0.2), main="Conditional frequencies on rows")
# zebraing 
pyramid(tab, gap=c(0.2, 0.2), 
        left=list(list(col="black"), list(col="white")), 
        right=list(list(col="blue"), list(col="green")))

resetpar

Description

Resets the par if necessary.

Usage

resetpar(oldpar)
resetpar(oldpar)

Arguments

oldpar

graphical parameters

Value

nothing

Examples

# no examples
# no examples

sandrews

Description

Shiny app for creating an Andrews curve diagram with interactive variable selection.

Usage

sandrews(data, xvar = character(0), ...)
sandrews(data, xvar = character(0), ...)

Arguments

`data`	matrix or data frame
`xvar`	character: names of selected variables for the plot
`...`	unused

Value

nothing

Examples

if (interactive()) sandrews(iris)
if (interactive()) sandrews(iris)

schernoff

Description

Shiny app for creating a Chernoff faces plot with interactive variable selection.

Usage

schernoff(data, xvar = character(0), ...)
schernoff(data, xvar = character(0), ...)

Arguments

`data`	matrix or data.frame
`xvar`	character: names of selected variables for the plot
`...`	further parameters given to DescTools::PlotFaces

Value

nothing

Examples

if (interactive()) schernoff(normalize(iris))
if (interactive()) schernoff(normalize(iris))

sdbscan

Description

Shiny app which allows to run a cluster analysis with DBSCAN with interactive choice of variables, core distance, and minimal neighbours.

Usage

sdbscan(data, xvar = character(0), ...)
sdbscan(data, xvar = character(0), ...)

Arguments

`data`	matrix or data.frame
`xvar`	character: names of selected variables for the clustering
`...`	unused

Value

nothing

Examples

if (interactive()) sdbscan(iris)
if (interactive()) sdbscan(iris)

sdistance

Description

Shiny app which shows the contribution of each variable to the distance between two observations with interactive variable selection. If $dijk$ is the distance between observations $i$ and $j$ in variable $k$ then the contribution is computed:

Usage

sdistance(data, xvar = character(0), ...)
sdistance(data, xvar = character(0), ...)

Arguments

`data`	matrix or data.frame
`xvar`	character: names of selected variables for the plot
`...`	unused

Details

Total variance: $var_k/sum(var_k)$ with $var_k$ the variance of the $k$ th variable
Minimum: $dijk==min_k(dijk)$
Manhattan: $dijk/sum(dijk)$
Gower: $dijk$ is rescaled to $[0, 1]$ in each variable and then $dijk/sum(dijk)$
Euclidean: $dijk^2/sum(dijk^2)$
Manhattan: $dijk/sum(dijk)$
Maximum: $dijk==max_k(dijk)$

Value

nothing

Examples

if (interactive()) sdistance(iris)
if (interactive()) sdistance(iris)

sfactor

Description

Shiny app for doing a factor analysis with interactive variable selection.

Usage

sfactor(data, xvar = character(0), ...)
sfactor(data, xvar = character(0), ...)

Arguments

`data`	matrix or data frame
`xvar`	character: names of selected variables for the plot
`...`	unused

Value

nothing

Examples

if (interactive()) sfactor(iris)
if (interactive()) sfactor(iris)

shclust

Description

Shiny app which allows to run a hierarchical cluster analysis with interactive choice of variables, distance, and agglomeration method.

Usage

shclust(data, xvar = character(0), ...)
shclust(data, xvar = character(0), ...)

Arguments

`data`	matrix or data.frame
`xvar`	character: names of selected variables for the clustering
`...`	unused

Value

nothing

Examples

if (interactive()) shclust(iris)
if (interactive()) shclust(iris)

skmeans

Description

Shiny app which allows to run a k-means cluster analysis with interactive choice of variables.

Usage

skmeans(data, xvar = character(0), ...)
skmeans(data, xvar = character(0), ...)

Arguments

`data`	matrix or data.frame
`xvar`	character: names of selected variables for the clustering
`...`	unused

Value

nothing

Examples

if (interactive()) skmeans(iris)
if (interactive()) skmeans(iris)

smclust

Description

Shiny app which allows to run a EM clustering with interactive choice of variables.

Usage

smclust(data, xvar = character(0), ...)
smclust(data, xvar = character(0), ...)

Arguments

`data`	matrix or data.frame
`xvar`	character: names of selected variables for the clustering
`...`	unused

Value

nothing

Examples

if (interactive()) smclust(iris)
if (interactive()) smclust(iris)

smosaic

Description

Shiny app for creating a Mosaic plot with interactive variable selection.

Usage

smosaic(data, xvar = character(0), yvar = character(0), ...)
smosaic(data, xvar = character(0), yvar = character(0), ...)

Arguments

`data`	table or data.frame
`xvar`	character: names of selected variables for x-axis
`yvar`	character: names of selected variables for y-axis
`...`	further parameters given to graphics::mosaicplot

Value

nothing

Examples

if (interactive()) smosaic(Titanic)
dfTitanic <- toDataframe(Titanic)
if (interactive()) smosaic(dfTitanic)
if (interactive()) smosaic(Titanic)
dfTitanic <- toDataframe(Titanic)
if (interactive()) smosaic(dfTitanic)

sortbin

Description

Sorts and bins the rows of the data frame x according the sorting columns in sortCol. decreasing and na.last are recycled is necessary. If equibin is TRUE and nBins==NA then nBins is set to 100. If equibin is FALSE and nBins==NA then the bins are returned as they come from sorting; only identical values are in one bin. If nBins is positive then the bins are merged until nBins reached. Note that the numbers of observations per bin may vary.

Usage

sortbin(
  x,
  sortCol = 1,
  decreasing = FALSE,
  na.last = TRUE,
  nBins = NA,
  equibin = TRUE
)
sortbin(
  x,
  sortCol = 1,
  decreasing = FALSE,
  na.last = TRUE,
  nBins = NA,
  equibin = TRUE
)

Arguments

`x`	data frame
`sortCol`	numeric/character: names or indices of variable used for sorting (default: `1`)
`decreasing`	logical: should the sort order be increasing or decreasing (default: `FALSE`)
`na.last`	logical: for controlling the treatment of NAs (default: `TRUE`)
`nBins`	integer: maximal number of bins (default: `NA`).
`equibin`	logical: should the number of observations equal per bin (default: `TRUE`)

Value

(non-sequential) bin numbers as integer

Examples

data("Boston", package="MASS")
tableplot(Boston, bin=sortbin(Boston))
data("Boston", package="MASS")
tableplot(Boston, bin=sortbin(Boston))

spairs

Description

Shiny app for creating a scatterplot matrix with interactive variable selection.

Usage

spairs(data, xvar = character(0), ...)
spairs(data, xvar = character(0), ...)

Arguments

`data`	matrix or data.frame
`xvar`	character: names of selected variables for the plot
`...`	further parameters given to graphics::pairs

Value

nothing

Examples

if (interactive()) spairs(iris)
if (interactive()) spairs(iris)

sandrews

Description

Shiny app for creating a Parallel Coordinate plot with interactive variable selection.

Usage

sparcoord(data, xvar = character(0), ...)
sparcoord(data, xvar = character(0), ...)

Arguments

`data`	matrix or data.frame
`xvar`	character: names of selected variables for the plot
`...`	further parameters given to MASS::parcoord

Value

nothing

Examples

if (interactive()) sparcoord(iris)
if (interactive()) sparcoord(iris)

splot

Description

Shiny app for choosing a specific plot.

Usage

splot(data, xvar = character(0), path = NULL)
splot(data, xvar = character(0), path = NULL)

Arguments

`data`	data.frame: input data
`xvar`	character: selected variables (default: `character(0)`)
`path`	character: path where to read the plot modules (default: `NULL`)

Value

nothing

Examples

if (interactive()) splot(iris)
if (interactive()) splot(iris)

sradar

Description

Shiny app for creating radar charts with interactive variable selection.

Usage

sradar(data, xvar = character(0), ...)
sradar(data, xvar = character(0), ...)

Arguments

`data`	matrix or data.frame
`xvar`	character: names of selected variables for the plot
`...`	unused

Value

nothing

Examples

if (interactive()) sradar(iris)
if (interactive()) sradar(iris)

tableplot

Description

A tableplot is a visualisation of multivariate data sets. Each column represents a variable and each row bin is an aggregate of a certain number of records. For numeric variables, a value box is plotted with minimum, mean (black line) and maximum value. If any missing values in a bin of a numeric variable appear the box left from the value box is plotted in gray. For categorical variables, a stacked bar chart is depicted of the proportions of categories. Missing values are taken into account.

Usage

tableplot(
  x,
  select = NULL,
  subset = NULL,
  bin = NULL,
  yj = NA,
  IQR_bias = 5,
  colpal = grDevices::rainbow,
  color.NA_num = "gray75",
  color.NA = "grey75",
  color.num = "lightblue",
  color.box = "deepskyblue",
  color.line = "black",
  box.lower = NULL,
  box.upper = NULL,
  box.line = NULL,
  cex.main = 1,
  cex.legend = 1,
  width = 1,
  height = 0.15
)
tableplot(
  x,
  select = NULL,
  subset = NULL,
  bin = NULL,
  yj = NA,
  IQR_bias = 5,
  colpal = grDevices::rainbow,
  color.NA_num = "gray75",
  color.NA = "grey75",
  color.num = "lightblue",
  color.box = "deepskyblue",
  color.line = "black",
  box.lower = NULL,
  box.upper = NULL,
  box.line = NULL,
  cex.main = 1,
  cex.legend = 1,
  width = 1,
  height = 0.15
)

Arguments

`x`	data frame
`select`	numeric/character: variable to show in the plot (default: `NULL`)
`subset`	numeric: index of observations to show
`bin`	integer: bin numbers to which a observations belongs (default: `NULL` = all)
`yj`	numeric: Yeo Johnson coefficient (default: `NA`). If `NA` then it will be set to 0 (=log) or 1 (=identity)
`IQR_bias`	numeric: parameter that determines when a logarithmic scale is used when `yj` is set to `NA`. The argument IQR_bias is multiplied by the interquartile range as a test.
`colpal`	color palette to draw (default: `rainbow`)
`color.NA_num`	color for missing of infinity values for numeric variables (default: `gray75`)
`color.NA`	color for missing values for categorical variables (default: `grey75`)
`color.num`	color for lower box for numeric variables (default: `lightblue`)
`color.box`	color for upper box for numeric variables (default: `deepskyblue`)
`color.line`	color for line in upper box for numeric variables (default: `black`)
`box.lower`	function: determine lower border in upper box for numeric variables (default: `NULL`). If `NULL` then `min(.,na.rm=TRUE)` is used.
`box.upper`	function: determine upper border in upper box for numeric variables (default: `NULL`). If `NULL` then `max(.,na.rm=TRUE)` is used.
`box.line`	function: determine line position in upper box for numeric variables (default: `NULL`). If `NULL` then `mean(.,na.rm=TRUE)` is used.
`cex.main`	number: magnification to be used for the titles (default: `1`)
`cex.legend`	number: magnification to be used for the legends (default: `1`)
`width`	number: width of percentage axis (default: `1`). If `1` then the width is as wide as a plot.
`height`	number: percentage of the height of the legends (default: `0.15`)

Details

The idea and some code of the tableplot is taken from tableplot package by Martijn Tennekes and Edwin de Jonge. It differs from their package by

multicolumn sorting is possible, and
no support for 'ff' (out of memory vectors).

Value

nothing

References

Tennekes, M., Jonge, E. de, Daas, P.J.H. (2013), Visualizing and Inspecting Large Datasets with Tableplots, Journal of Data Science 11 (1), 43-58.

Examples

data("Boston", package="MASS")
tableplot(Boston, bin=sortbin(Boston))
data("Boston", package="MASS")
tableplot(Boston, bin=sortbin(Boston))

template

Description

Each line of a code template consists of condition based on the unnamed parameters and R code in which replacements with named parameters done.

Usage

template(text, ...)
template(text, ...)

Arguments

`text`	a code template
`...`	further parameters

Value

a character vector

Examples

template("
1:   'Hello {{letter}}'
!1:  'Good-bye {{letter}}'
         ",
         letter=sample(LETTERS, 1),
         runif(1)<0.5 #1 = first unnamed parameter
         )
template("
1:   'Hello {{letter}}'
!1:  'Good-bye {{letter}}'
         ",
         letter=sample(LETTERS, 1),
         runif(1)<0.5 #1 = first unnamed parameter
         )

Test data

Description

A data frame containing various variable types and special values.

Usage

testdata
testdata

Format

A data frame with n=25 rows and 8 variables:

xu: runif(n) with a NA, NaN, Inf, -Inf
xn: rnorm(n, 0, 2) with a NA, NaN
x0: rep(0, n)
xi: ⁠as.integer(rnorm(n, 0, 2)⁠ with a NA, NaN
x2: sample(c(0,1), size=n, replace=TRUE)
gf: ⁠factor(as.integer(rnorm(n, 0, 2))⁠ with a NA
go: ⁠ordered(as.integer(rnorm(n, 0, 2))⁠ with a new level 10
gn: ⁠ordered(as.integer(rnorm(n, 0, 2))⁠ with a NA
gc: ⁠as.character(as.integer(rnorm(n, 0, 2))⁠ with a NA and ""
gl: ⁠sample(c(T,F), size=n, replace=TRUE))⁠ with a NA
g0: ⁠rep("constant, n)⁠
g2: sample(c(T,F), size=n, replace=TRUE)

toChoice

Description

The elements in ... will coerced into one text vector. The entries will either the text (method==NA) or integer number starting at method. The first letter of the list element names will be capitalized.

Usage

toChoice(method = NA, ...)
toChoice(method = NA, ...)

Arguments

`method`	integer: which method is used for creating the list elements
`...`	character: choice values

Value

a list

Examples

txt <- c("the", "quick", "brown", "fox", "jumps", "over", "the", "lazy", "dog")
toChoice(NA, txt)
toChoice(0, txt) # integer sequence starts at zero
txt <- c("the", "quick", "brown", "fox", "jumps", "over", "the", "lazy", "dog")
toChoice(NA, txt)
toChoice(0, txt) # integer sequence starts at zero

toDataframe

Description

Converts a table to a full data frame.

Usage

toDataframe(obj, name = NULL, ...)
toDataframe(obj, name = NULL, ...)

Arguments

`obj`	R object (`table` or `ts`) to convert to a data frame
`name`	character: vector of variable name(s), only use for a `ts` object
`...`	further parameters given to base::as.data.frame.table

Value

a data frame

Examples

toDataframe(Titanic)
toDataframe(austres)
toDataframe(Titanic)
toDataframe(austres)

toLayout

Description

Given a (minimal) length len and the number unused entries a layout is generated. If sel<0 then less rows and more columns are used and if sel>0 then more rows and less columns are used.

Usage

toLayout(len, sel = 0, unused = 0)
toLayout(len, sel = 0, unused = 0)

Arguments

`len`	integer: minimal size of layout
`sel`	integer: select less or more rows (default: `0`)
`unused`	integer: number of unused entries (default: `0`)

Value

a matrix

Examples

toLayout(13)
toLayout(13)

toRDS

Description

Saves one or more data sets in RDS format to a temporary directory (tmpdir()). Data sets must have the class ts or something that can be converted to a data frame, e.g. matrix, table, etc.

Usage

toRDS(...)
toRDS(...)

Arguments

...

data sets to save

Value

returns the name of the created files

Examples

toRDS(Titanic) # saves to tempdir/Titanic.rds
toRDS(Titanic) # saves to tempdir/Titanic.rds

Trend and aeasonality estimation of a univariate time series

Description

Estimate a trend and seasonaliyt for a time series. Available functions:

trend_season to generate an estimate
print to print the estimate
summary to summarize the etsimate result
plot to plot the time series, its estimation and the residuals
coef to extract the coefficients if a seasonality estimation was done
residuals to extract the residuals of the model
fitted to the fitted values

Usage

trend_season(t, ...)

## Default S3 method:
trend_season(
  t,
  trend = c("constant", "linear", "exponential"),
  season = c("none", "additive", "multiplicative"),
  ...
)

## S3 method for class 'trend_season'
print(x, ...)

## S3 method for class 'trend_season'
summary(object, ...)

## S3 method for class 'trend_season'
plot(x, y, which = 1, ...)
trend_season(t, ...)

## Default S3 method:
trend_season(
  t,
  trend = c("constant", "linear", "exponential"),
  season = c("none", "additive", "multiplicative"),
  ...
)

## S3 method for class 'trend_season'
print(x, ...)

## S3 method for class 'trend_season'
summary(object, ...)

## S3 method for class 'trend_season'
plot(x, y, which = 1, ...)

Arguments

`t`	ts: time series object
`...`	unused
`trend`	character: trend method, either `none` (default), `linear` or `exponential`
`season`	character: seasonality method, either `none` (default), `additive` or `multiplicative`
`x`, `object`	trend_season: estimated time series
`y`	unused
`which`	integer: what to plot, `1` time series and estimation (default) or `2` residuals

Value

trend_season returns a trend_season object with

call the function call
ts the input time series
trend the trend estimation (ts object)
trend.residuals the residuals of the trend estimation (ts object)
season the trend and season estimation (ts object)
season.residuals the residuals of the trend and season estimation (ts object)
coefficients the coefficients used in the seasonality estimation
residuals the residuals of the model
fitted.values the fitted values of the model

Examples

tts <- trend_season(austres, "linear")
print(tts)
summary(tts)
plot(tts)
plot(tts, which=2)
residuals(tts)
fitted(tts)
coef(tts)  # if NULL then no seasonality was estimated
tts <- trend_season(austres, "linear")
print(tts)
summary(tts)
plot(tts)
plot(tts, which=2)
residuals(tts)
fitted(tts)
coef(tts)  # if NULL then no seasonality was estimated

General UI elements

Description

Some general UI elements for common use where last selected value is stored for reuse:

UIplottype plot type, defines smvgraph_type
UIpointsymbol plot symbol for point, defines smvgraph_pch
UIpointsize point size, defines smvgraph_cex
UIlinetype line type, defines smvgraph_lty
UIlinewidth line width, defines smvgraph_lwd
UItextsize text size, defines smvgraph_tex
UIlegend legend position, defines smvgraph_legend
UIlegendsize legend size, defines smvgraph_lex
UIdatanormalization should data be rescaled, defines smvgraph_normalize (no, minMax, mtandardization)
UIdistance distance to use, defines smvgraph_distance
UIobservations range of observations, defines smvgraph_obs
UImergegroups should a set of grouping variables merged into one group variable, defines smvgraph_single

From the top menu are the following input elements are defined

input$smvgraph_pch point symbol,
input$smvgraph_cex point size,
input$smvgraph_lty line type,
input$smvgraph_lwd line width,
input$smvgraph_tex text size, and
inpus$smvgraph_legend legend position.

Usage

UIdatanormalization(
  sel = getShinyOption("smvgraph.current")$smvgraph_normalize
)

UIdistance(sel = getShinyOption("smvgraph.current")$smvgraph_distance)

UIobservations(n, sel = getShinyOption("smvgraph.current")$smvgraph_obs)

UImergegroups(n, sel = getShinyOption("smvgraph.current")$smvgraph_single)

UIpointsize(n, sel = getShinyOption("smvgraph.current")$smvgraph_cex)

UIpointsymbol(n, sel = getShinyOption("smvgraph.current")$smvgraph_pch)

UIlinewidth(n, sel = getShinyOption("smvgraph.current")$smvgraph_lwd)

UIlinetype(n, sel = getShinyOption("smvgraph.current")$smvgraph_lty)

UItextsize(n, sel = getShinyOption("smvgraph.current")$smvgraph_tex)

UIlegend(n, sel = getShinyOption("smvgraph.current")$smvgraph_legend)

UIlegendsize(n, sel = getShinyOption("smvgraph.current")$mvgraph_lex)

UIplottype(n, sel = getShinyOption("smvgraph.current")$smvgraph_type)
UIdatanormalization(
  sel = getShinyOption("smvgraph.current")$smvgraph_normalize
)

UIdistance(sel = getShinyOption("smvgraph.current")$smvgraph_distance)

UIobservations(n, sel = getShinyOption("smvgraph.current")$smvgraph_obs)

UImergegroups(n, sel = getShinyOption("smvgraph.current")$smvgraph_single)

UIpointsize(n, sel = getShinyOption("smvgraph.current")$smvgraph_cex)

UIpointsymbol(n, sel = getShinyOption("smvgraph.current")$smvgraph_pch)

UIlinewidth(n, sel = getShinyOption("smvgraph.current")$smvgraph_lwd)

UIlinetype(n, sel = getShinyOption("smvgraph.current")$smvgraph_lty)

UItextsize(n, sel = getShinyOption("smvgraph.current")$smvgraph_tex)

UIlegend(n, sel = getShinyOption("smvgraph.current")$smvgraph_legend)

UIlegendsize(n, sel = getShinyOption("smvgraph.current")$mvgraph_lex)

UIplottype(n, sel = getShinyOption("smvgraph.current")$smvgraph_type)

Arguments

`sel`	selected element
`n`	integer: number of observations

Value

an UI element for shiny

Examples

# none
# none

valid, invalid

Description

Computes the number a logical matrix or vector if all values are valid in x or each colum or row of x. Valid values for numeric variables are is.finite(v) and for other types !is.na

Computes the number a logical matrix or vector if any values are valid in x or each colum or row of x.

Usage

valid(x, margin = 1:2, n = FALSE)

invalid(x, margin = 1:2, n = FALSE)
valid(x, margin = 1:2, n = FALSE)

invalid(x, margin = 1:2, n = FALSE)

Arguments

`x`	object: anything taht can be coerced to a data frame checked for valid/invalid values
`margin`	integer: a vector giving the subscripts for which valid/invalid values looked for, e.g. `margin==1` indicates rows, `margin==2` indicates columns, otherwise indicates rows and columns.
`n`	logical: should just the number of valid/invalid values returned or a logical matrix/vector

Value

a logical data frame, a logical vector or an integer

Examples

data("testdata")
valid(testdata)             # matrix with logical entries if x has valid entry
valid(testdata, n=TRUE)     # number of valid entries in x
valid(testdata, 1)          # vector with logical entries if each row if x has valid entries
valid(testdata, 1, n=TRUE)  # number of rows with valid entries in x
valid(testdata$xu)
data("testdata")
valid(testdata)             # matrix with logical entries if x has valid entry
valid(testdata, n=TRUE)     # number of valid entries in x
valid(testdata, 1)          # vector with logical entries if each row if x has valid entries
valid(testdata, 1, n=TRUE)  # number of rows with valid entries in x
valid(testdata$xu)

Progress

Description

Reports progress to the user during long-running operations.

Usage

with_progress(...)

set_progress(...)

inc_progress(...)
with_progress(...)

set_progress(...)

inc_progress(...)

Arguments

...

see [shiny::withProgress]

Value

see [shiny::withProgress]

Examples

## Only run examples in interactive R sessions
if (interactive()) {
   options(device.ask.default = FALSE)
   ui <- fluidPage(plotOutput("plot"))
   #
   server <- function(input, output) {
     output$plot <- renderPlot({
       with_progress(message = 'Calculation in progress',
                     detail = 'This may take a while...', value = 0, {
                     for (i in 1:15) {
                       inc_progress(1/15)
                       Sys.sleep(0.25)
                     }
                   })
       plot(cars)
   })
 }
 #
 shinyApp(ui, server)
}
## Only run examples in interactive R sessions
if (interactive()) {
   options(device.ask.default = FALSE)
   ui <- fluidPage(plotOutput("plot"))
   #
   server <- function(input, output) {
     output$plot <- renderPlot({
       with_progress(message = 'Calculation in progress',
                     detail = 'This may take a while...', value = 0, {
                     for (i in 1:15) {
                       inc_progress(1/15)
                       Sys.sleep(0.25)
                     }
                   })
       plot(cars)
   })
 }
 #
 shinyApp(ui, server)
}

yeo.johnson

Description

Computes the Yeo-Johnson transformation, which is a normalizing transformation. The code and documentation is taken from the VGAM package (see function yeo.johnson) with some slight modifications, e.g. NA's are kept and do not produce an error.

Usage

yeo.johnson(
  y,
  lambda,
  derivative = 0,
  epsilon = sqrt(.Machine$double.eps),
  inverse = FALSE
)
yeo.johnson(
  y,
  lambda,
  derivative = 0,
  epsilon = sqrt(.Machine$double.eps),
  inverse = FALSE
)

Arguments

`y`	numeric: a vector or matrix.
`lambda`	numeric: It is recycled to the same length as `y` if necessary.
`derivative`	non-negative integer: the default is the ordinary function evaluation, otherwise the derivative with respect to `lambda` (default: `0`)
`epsilon`	numeric and positive value: the tolerance given to values of `lambda` when comparing it to 0 or 2.
`inverse`	logical: return the inverse transformation? (default: `FALSE`)

Details

The Yeo-Johnson transformation can be thought of as an extension of the Box-Cox transformation. It handles both positive and negative values, whereas the Box-Cox transformation only handles positive values. Both can be used to transform the data so as to improve normality.

Value

The Yeo-Johnson transformation or its inverse, or its derivatives with respect to lambda, of y.

Note

If inverse = TRUE then the argument derivative = 0 is required.

References

Yeo, I.-K. and Johnson, R. A. (2000). A new family of power transformations to improve normality or symmetry. Biometrika, 87, 954–959.

Examples

y <- seq(-4, 4, len = (nn <- 200))
ltry <- c(0, 0.5, 1, 1.5, 2)  # Try these values of lambda
lltry <- length(ltry)
psi <- matrix(as.numeric(NA), nn, lltry)
for (ii in 1:lltry)
  psi[, ii] <- yeo.johnson(y, lambda = ltry[ii])
matplot(y, psi, type = "l", ylim = c(-4, 4), lwd = 2, lty = 1:lltry,
        ylab = "Yeo-Johnson transformation", col = 1:lltry, las = 1,
        main = "Yeo-Johnson transformation with some values of lambda")
abline(v = 0, h = 0)
legend(x = 1, y = -0.5, lty = 1:lltry, legend = as.character(ltry),
       lwd = 2, col = 1:lltry)
y <- seq(-4, 4, len = (nn <- 200))
ltry <- c(0, 0.5, 1, 1.5, 2)  # Try these values of lambda
lltry <- length(ltry)
psi <- matrix(as.numeric(NA), nn, lltry)
for (ii in 1:lltry)
  psi[, ii] <- yeo.johnson(y, lambda = ltry[ii])
matplot(y, psi, type = "l", ylim = c(-4, 4), lwd = 2, lty = 1:lltry,
        ylab = "Yeo-Johnson transformation", col = 1:lltry, las = 1,
        main = "Yeo-Johnson transformation with some values of lambda")
abline(v = 0, h = 0)
legend(x = 1, y = -0.5, lty = 1:lltry, legend = as.character(ltry),
       lwd = 2, col = 1:lltry)

zzz

Description

Runs all s... functions for test purposes if interactively called.

Usage

zzz()
zzz()

Value

nothing

Examples

zzz()
zzz()

Package 'smvgraph'

Help Index

andrews

Description

Usage

Arguments

Value

References

See Also

Examples

as_param

Description

Usage

Arguments

Value

Examples

availablePlots

Description

Usage

Details

Value

Examples

bagplot

Description

Usage

Arguments

Details

Value

References

See Also

Examples

binData

Description

Usage

Arguments

Value

Examples

character_data

Description

Usage

Arguments

Value

Examples

checkPackages, installPackages

Description

Usage

Arguments

Value

Examples

color_data

Description

Usage

Arguments

Value

Examples

color_hclust

Description

Usage

Arguments

Value

Examples

convertTo

Description

Usage

Arguments

Value

Examples

factor_data

Description

Usage

Arguments

Value

Examples

formatCommands

Description

Usage

Arguments

Value

Examples

getModules