A collection of tools that the author finds handy

Description

Most of these tools are small functions to do simple tasks or provide filtered views of the current environment. In addition the function splatnd is provided primarily as a piece of example code to show how to write zero-argument operators. It's based on the code in the package sos , and avoids the need to use makeActiveBinding (as in ,e.g., pracma::ans )

Author(s)

Carl Witthoft, with attributions as noted in the individual help pages

Maintainer:Carl Witthoft carl@witthoft.com

Do "fuzzy" equality and return a logical vector.

Description

This function compares two vectors (or arrays) of values and returns the near-equality status of corresponding elements. As with all.equal() , the intent is primarily to get around machine limits of representation of floating-point numbers. For integer comparison, just use the base == operator.

Usage

approxeq(x, y, tolerance = .Machine$double.eps^0.5, ...)

Arguments

Details

If x and y are of different lengths, the shorter one is recycled and a warning issued.

Value

A vector of the same length as the longer of x or y, consisting of TRUE and FALSE elements, depending on whether the corresponding elements of x and y are within the approximate equality precision desired.

Author(s)

Carl Witthoft, carl@witthoft.com

See Also

all.equal, Comparison, identical

Examples

x<-1:10
y<-x+runif(10)*1e-6
approxeq(x,y) #all FALSE
approxeq(x,y,tolerance=1e-5) #mostly TRUE, probably

Banner Versions of Characters

Description

This data set contains the "banner" arrays of characters to be used by the function banvax.

Usage

ascarr

Format

The array must be of dimension [6,8,N] for N characters. The dimnames for the third dimension must be the character matching each such layer.

Interactive application of selected function a list of objects.

Description

This function was originally written to do the same as the unix rm -i command. The user supplies a list of items and the name of a function which is optionally applied to each item in turn.

Usage

askrm(items = ls(parent.frame()), fn = "rm", ask = TRUE)

Arguments

Value

A list with three elements.

Author(s)

Carl Witthoft, carl@witthoft.com

See Also

When interactive prompting is not desired, sapply or its brethren are recommended.

Examples

# get rid of junky objects left around from testing
foo<-1
afoo<-c(foo,2)
foob <- c('a','b','d')
askrm(ls(pattern="foo") )

x<- rep(1,10)
y<- runif(10)
askrm(c('x','y'),'sd',ask=FALSE)

Function print a text banner in the style of the original DEC VAX "banner" function.

Description

This function takes a character string and writes the same characters as large block-letters into a text file.

Usage

 
banvax(msg, file = 'banner.txt',  linewid = 80, bandat = cgwtools::ascarr)

Arguments

Details

The supplied data file ascarr contains all letters and numerals and a bunch of punctuation marks such as " ;" , "," "\" , etc. If a character in the input is not found in the file, a "box" is used in its place. If a user-supplied file is specified, note that each layer of the [x,y,N] array must have a name equal to the character to be invoked. Here x specifies the number of columns and y the number of rows in each banner-element.

The function cat generates the output, so setting the argument file to "" will direct the output to the console. Quoting from the help page for cat, ' If "" (the default), cat prints to the standard output connection, the console unless redirected by sink. If it is "|cmd", the output is piped to the command given by cmd , by opening a pipe connection. '

Value

Nothing is returned from the function. The output is a file or whatever connection is specified by the file argument.

Author(s)

Author and Maintainer:Carl Witthoft carl@witthoft.com

References

https://www0.mi.infn.it/~calcolo/OpenVMS/ssb71/6015/6017p041.htm#index_x_2757

Function to convert any base to any other base (up to 36).

Description

This function now resides in the package base2base

Create histogram bins for each unique value in a sample.

Description

This is a Q&D way to create Pareto / histogram bins of a dataset when you want a separate bin for each value and don't want to deal with the 'breaks' or equivalent arguments in hist or other histogram functions in R packages.

Usage

binit(samps,roundPrec=NULL)

Arguments

Details

binit sorts the input data and feeds the result to rle. This effectively produces histogram-like results. If you want a strict Pareto order (most common first), just sort the list elements lengths and values by the magnitudes in lengths.

Value

A list containing two elements lengths: the number of items in each bin values: the data value associated with each bin

Author(s)

Carl Witthoft, carl@witthoft.com

See Also

rle , hist,

Examples

x <- sample(1:100, 1000, rep=TRUE)
xbin <- binit(x)
plot(xbin$values,xbin$lengths, type = 'h')
# without rounding, will just be grass
x <- rnorm(1000)
 xbin <- binit(x,2)
plot(xbin$values,xbin$lengths, type = 'h')

Function calculate the cumulative result of any function along an input vector.

Description

Calculates the cumulative value of almost any function in the same manner as cumsum

Usage

cumfun(frist, FUN = sum, ...)

Arguments

Details

If additional arguments are of length one, they are applied to each value in frist. If they are longer but not equal in length to length(frist) they will either be truncated or recycled to match. The builtin functions cumsum and cumprod generally are much faster than applying sum or prod to this function.

Value

A list of the cumulative calculations generated.

Author(s)

Carl Witthoft, carl@witthoft.com

See Also

cumsum , cumall, rowCumsums

Examples

foo <- 'abcdefcghicklmno'
grepfoo <-function(x,y) grep(y,x)
cumfoo <- cumfun(unlist(strsplit(foo,'')), FUN=grepfoo, y = 'c')
bar <- rnorm(1000,1)
cumsd <- cumfun(bar,FUN=sd)
plot(unlist(cumsd),type='l')
# compare with input std dev
lines(c(0,1000),c(1,1),lty=2,col='red')

Function to return dimensions of arguments, or lengths if dim==NULL.

Description

Simple overload to return dim when it's sensible and length otherwise

Usage

dim(item)

Arguments

Value

Either a single value as returned by length or a vector of integers indicating the magnitude of each dimension as returned by dim

Author(s)

Carl Witthoft, carl@witthoft.com

See Also

dim , length,

Examples

x1<-1:10
x2<-matrix(1,3,4)
dim(x1)
dim(x2)

Wrapper function around dir() which returns only directories found in the specified location(s).

Description

For those times when you only want to know the local directories available, use this instead of struggling through myriad arguments to dir . All arguments are the same as for plain old "dir" and are passed to dir .

Usage

dirdir(path = ".", pattern = NULL, all.files = FALSE, full.names = FALSE,
 recursive = FALSE, ignore.case = FALSE, include.dirs = FALSE, no.. = FALSE)

Arguments

Value

Note: this is directly quoted from the man page for dir. A character vector containing the names of the files in the specified directories (empty if there were no files). If a path does not exist or is not a directory or is unreadable it is skipped.

The files are sorted in alphabetical order, on the full path if full.names = TRUE.

Author(s)

Carl Witthoft, carl@witthoft.com

See Also

dir

Function to locate patterns ( sequences of strings or numerical values) in data vectors.

Description

Finds the location of a specified sequence either of numbers or strings in the source data item. If desired, for numerical data, both the source and the named pattern can be rounded to specified number of digits.

Usage

findpat(datavec, pattern, roundit = NULL)

Arguments

Details

If datavec and pattern are of different types, R will automagically convert to a common type and then compare the values. This is a result of the coercion rules identified in the man page for Comparison, "If the two arguments are atomic vectors of different types, one is coerced to the type of the other, the (decreasing) order of precedence being character, complex, numeric, integer, logical and raw." Use of the roundit argument is recommended whenever working with doubles (floats) to avoid the well-known and often overlooked pain of binary precision errors.

Value

If the first element of pattern isn't found, a message is posted and an empty integer vector is returned. Otherwise, a vector of the indices of datavec where the desired pattern is located is returned. These are the indices of the start of the pattern.

Author(s)

Carl Witthoft, carl@witthoft.com

See Also

which , nth_number_after_mth, strfind

Examples

fooc <- letters[c(1:15,4:9,12:26)]
 findpat(fooc,c('d','e','f'))
# 4 16
fooi <- c(1:50,5:9,60:80)
findpat(fooi,6:8)
# 6 52
findpat(fooi,c('6','7','8'))
# also 6 52

Returns the current directory stack that pushd and popd manipulate

Description

getstack goes into the separate environment where pushd and popd operate and returns the current stack of directories.

Usage

getstack()

Arguments

none

Details

Allowing a function to modify an object in the GlobalEnvironment is frowned upon by CRAN (and most programmers), so to maintain a directory stack a separate environment is established by pushd. Since this environment is not visible at the console level, getstack allows the user to check on the current status of the stack.

Value

The current directory stack is returned as a vector of strings.

Author(s)

Carl Witthoft carl@witthoft.com

See Also

popd , pushd, setwd

Examples

## depends on your local directory structure and permissions
getwd()
getstack() #empty, probably
pushd('..')
getstack()
pushd('.')
getstack()
popd()
getstack()
popd()
getstack()
getwd() #back where we started

Inverse of seqle

Description

As with inverse.rle, this function reverses the compression performed with seqle so long as you know the incr value used to generate the compressed data.

Usage

inverse.seqle(x, incr = 1L)

Arguments

Value

a vector of values identical (or nearly so, for floats) to the original sequence. Note: Since the concept of "increment" has no reliable meaning when dealing with characters or char strings, when x is non-numeric the argument incr is ignored and the function reverts to base::inverse.rle....

Note

The bulk of the code is taken directly from base::inverse.rle . Thanks to "flodel" on StackOverflow for suggesting code to handle floating-point increments.

Author(s)

Carl Witthoft, carl@witthoft.com

See Also

seqle , inverse.rle

Examples

x<- c(2,2,2,3:8,8,8,4,4,4,5,5.5,6)
y<-seqle(x,incr=0)
inverse.seqle(y,0)
y <- seqle(x,incr=1)
inverse.seqle(y)
inverse.seqle(y,2) # not what you wanted

Q&D function to list all objects with the specified class attribute .

Description

This is one of the author's collection of ls* Q&D functions. Since anyone can define a new class at any time, there is no predefined set of legal or illegal class names. Remember that an object can have multiple classes. This function only allows searching for a single class name in a given call.

Usage

lsclass(type = "numeric")

Arguments

Value

A vector of character strings containing the names of matching objects (as would be returned by the base function ls ) .

Author(s)

Carl Witthoft carl@witthoft.com

See Also

typeof, class , lstype

Examples

xyzzy<-structure(vector(),class='grue')
lsclass('integer')
lsclass('grue') 

List all objects in an .Rdata file.

Description

This function opens an .Rdata file, lists the contents, and cleans up after itself.

Usage

lsdata(fnam = ".Rdata")

Arguments

Value

The output of ls applied to the objects loaded from the specified data file.

Author(s)

Carl Witthoft carl@witthoft.com

References

Various people have published similar code on Stack Overflow.

See Also

load, resave

Examples

##not run because of complaints about detritus
# xblue<-1
# yblue<-2
# save(xblue,yblue,file='blue.Rdata')
# lsdata('blue.Rdata')

List the sizes of all selected objects.

Description

Just a toy to list the number of elements or optionally the bytesize as produced with object.size of a specified selection of objects. I find it handy when I want to rid an environment of large (or empty) objects. In the default case, byte=FALSE , lists and S4 objects are "taken apart" down to the lowest level so all individual elements are counted.

Usage

lssize(items, byte = FALSE)

Arguments

Value

A vector of the object sizes, with the object names as names for the elements

Author(s)

Carl Witthoft, carl@witthoft.com

References

Many thanks to Martin Morgan of bioconductor.org who provided the recursive function for deconstructing an S4 Object. See http://stackoverflow.com/questions/14803237/ for the original question and answer.

See Also

lstype , object.size, length

Examples

x1<-runif(100)
x2<-runif(1000)
x3<-runif(2000)
lssize(ls(pattern='x[1-3]'))
lssize(ls(pattern='x[1-3]'),byte=TRUE) 
#depending on what you have in your environment:
lssize(lstype('integer'))

List all objects of the specified type.

Description

This is a Q&D tool to list all objects in the current environment of a specified type. As discussed in the base R documentation, these types are the vector types "logical", "integer", "double", "complex", "character", "raw" and "list", "NULL", "closure" (function), "special" and "builtin" (basic functions and operators), "environment", "S4" (some S4 objects).

Usage

lstype(type = "closure")

Arguments

Value

A vector of character strings as is returned by the base function ls .

Author(s)

Carl Witthoft carl@witthoft.com

See Also

ls , lssize, lsclass

Examples

lstype('integer') #if you have any such in your environment.

Functions to find the n-th maximum or minimum of a vector or array.

Description

These functions behave similarly to min, max and which(x == max/min (x) to find the n-th greatest max or min of a dataset.

Usage

maxn(x,nth=1)
minn(x,nth=1)
which.maxn(x, nth = 1, arr.ind = FALSE, useNames = TRUE)
which.minn(x, nth = 1, arr.ind = FALSE, useNames = TRUE) 

Arguments

Details

Quoting the help page for max : Character versions are sorted lexicographically, and this depends on the collating sequence of the locale in use: the help for Comparison gives details. The max/min of an empty character vector is defined to be character NA. (One could argue that as "" is the smallest character element, the maximum should be "", but there is no obvious candidate for the minimum.)

Value

For maxn, minn, a single value which is the nth- max or min.

For the which.min,max functions, quoting from which: If arr.ind == FALSE (the default), an integer vector, or a double vector if x is a long vector, with length equal to sum(x), i.e., to the number of TRUEs in x. If arr.ind == TRUE and x is an array (has a dim attribute), the result is arrayInd(which(x), dim(x), dimnames(x)), namely a matrix whose rows each are the indices of one element of x

Author(s)

Carl Witthoft, carl@witthoft.com

See Also

which , max

Examples

set.seed(17) # for repeatability
foo <- matrix(sample(1:10,20,replace=TRUE),5,4)
maxn(foo,3)
which.minn(foo,4)

Functions which mimic max.col to find for minimum or maximum of rows or columns.

Description

These are Q&D wrappers around max.col to make it easy to get the positions of max or the min of either rows or columns of an array. The description of the base function is, for comparison, "Find the maximum position for each row of a matrix".

Usage

maxRow(mat,ties.method = c("random", "first", "last") )
minRow(mat,ties.method = c("random", "first", "last") )
minCol(mat,ties.method = c("random", "first", "last") )
maxCol(mat,ties.method = c("random", "first", "last") )

Arguments

Value

For each of these functions, same as for max.col : index of a max or min value for each row or column, an integer vector of length nrow(mat) or ncol(mat).

Author(s)

Carl Witthoft, carl@witthoft.com

See Also

max.col

Calculate and display basic statistics for an object.

Description

This function calculates the min, max, median, mean, standard deviation, skew and kurtosis for the specified object and displays the results in a semi-tabular form. An option is provided to set the number of digits displayed for the returned values. Note: see the help pages in this package for theskew and thekurt for information on those implementations.

Usage

mystat(x, numdig = 3, na.rm = TRUE, printit = TRUE)

Arguments

Value

A data frame with scalar elements matching their names:

Author(s)

Carl Witthoft, carl@witthoft.com

See Also

theskew , thekurt

Examples

x <- runif(100)
mystat(x)
mystat(x,numdig=6)

Function to find intersection points of two polygons.

Description

This is a Q&D tool to find the locations where two polygons, in a plane only (not 3D space), intersect.

Usage

polyInt(poly1,poly2, stopAtFirst = FALSE, plotit = FALSE, roundPrecision = 10, ...)

Arguments

Details

The function loops over all pairs of segments (one from poly1 and one from poly2), calling segSegInt to see if they intersect. After all pair-combinations are tested, the collected intersection points, if any, are reduced to the unique collection. This avoids repetition when an intersection point is a vertex of one (or both) of the polygons. It is not necessary to "close" the supplied set of vertices, i.e. repeat the initial vertex at the end of the array as is needed to generate a complete line-plot of a polygon. The function will add that repeated vertex if it's not present in the input polygon(s).

Note: The supporting function segSegInt returns NA when two segments are parallel. However, when two polygons in fact have an overlapping (and thus parallel) couple of edges, the adjoining edges of one or both polygons will not be parallel to these parallel edges and will intersect one or both, so the vertex which lies on the other polygon's edge will be reported.

Value

A matrix of the x and y coordinates of all intersection points, or, if stopAtFirst is TRUE, the first intersection point found. If no intersections exist, NULL is returned.

Author(s)

Carl Witthoft, carl@witthoft.com

See Also

There are many tools which are far faster and more flexible. I wrote this one because it uses only base functions and doesn't require converting polygon vertices into a special class variable. Here are two common packages. intersect , st_intersection

Examples

sqone <- cbind(c(0,1,1,0),c(0,0,1,1))
sqtwo <- sqone + 0.5 
foo <- polyInt(sqone, sqtwo, plotit = TRUE)

Performs equivalent of bash command with same name

Description

popd is based on the cygwin bash manpages' description of these commands.

Usage

popd(dn=FALSE, pull=0)

Arguments

Details

Recommend reading man bash for full details of the operations. This implementation will not change the working directory if dn is TRUE The directory history is stored in a file in the function's environment (not console environment).dirhist, typically first created with pushd .

Value

A status value: 0 for success or 1 if there is no stack file (.dirhist). Future upgrades may include other codes for other failure mechanisms, but for now error messages will have to suffice.

Author(s)

Carl Witthoft carl@witthoft.com

See Also

pushd , setwd

Examples

## depends on your local directory structure and permissions
getwd()
pushd("~/..")
getwd()
popd()
getwd()

Performs equivalent of bash command with same name

Description

pushd is based on the cygwin bash manpages' description of these commands.

Usage

pushd(path, dn=FALSE,rot=0)

Arguments

Details

Recommend reading man bash for full details of the operations. This implementation should do nothing more than change the working directory (and store directory history in a file in the function's environment (not console environment) .dirhist ).

Value

A status value, which is always 0 for success. A future upgrade may implement a trycatch for conditions such as an inaccessible directory, but for now error messages will have to suffice.

Author(s)

Carl Witthoft carl@witthoft.com

See Also

popd , setwd

Examples

## depends on your local directory structure and permissions
getwd()
pushd("~/..")
getwd()
popd()
getwd()

Function to find the rational roots of any polynomial (when they exist).

Description

.

Usage

ratRoot(Anum, Adenom = rep(1,times=length(Anum)) )

Arguments

Details

The code makes use of the 'Rational Root Theorem,' which states that all real, rational roots p/q meet two criteria. Given a polynomial of the form An *x^n + A_(n-1) * x^(n-1) + ... + A0 = 0, all coefficients adjusted to be integers, 'p' must be a factor of the zero-power polynomial coefficient A0, and 'q' must be a factor of the maximum-power polynomial coefficient An. If any of the input Adenom are not '1', the function will make this adjustment.

Value

A vector of bigq fractions representing all real rational roots found. If empty, all roots are irrational (or perhaps complex).

Author(s)

Carl Witthoft, carl@witthoft.com

References

https://en.wikipedia.org/wiki/Rational_root_theorem

Examples

ratRoot(c(6, -15, 261), c(1,1,49))
ratRoot(c(1,0,0,0,-1))
ratRoot(c(1,-2,1))

Add some objects to an existing .Rdata - type file.

Description

Take an existing myfile.Rdata data file and add the specified objects to it. This is achieved by opening the data file in a local environment, "dumping" the new objects into that environment, and re-saving everything to the same file name.

Usage

resave(..., list = character(), file, overwrite = TRUE, loadverbose = FALSE )

Arguments

Value

Nothing is returned. This function is used solely to put objects into the file.

Note

Code is essentially the same as that provided by "flodel" on StackOverflow, with some enhancements based on "save_append" by "nehartj" on GitHub

Author(s)

Carl Witthoft carl@witthoft.com

See Also

lsdata , save, load

Examples

## not run to avoid creating detritus
# foo<-1:4
# bar<-5:8
# save(foo,file='foo.Rdata')
# resave(bar,file='foo.Rdata')
# #check your work
# lsdata('foo.Rdata')

Function to find intersection point between two line segments (NOT lines).

Description

This function finds the intersection of the two lines containing the provided line segments, then checks whether the intersection is contained within both segments. This is an implementation of the equations presented, among many other sources, at https://paulbourke.net/geometry/pointlineplane/ . Bourke comments there, that "The equations apply to lines, if the intersection of line segments is required then it is only necessary to test if ua and ub lie between 0 and 1. Whichever one lies within that range then the corresponding line segment contains the intersection point. If both lie within the range of 0 to 1 then the intersection point is within both line segments."

Usage

segSegInt(seg1, seg2=NULL, plotit=FALSE, probParallel = 1e-10,  ...)

Arguments

Details

The function runs a check for parallelism so as not to return an infinity of intersection points, then basically checks for intersection of the lines to which the line segments belong, and finally verifies said intersection belongs to both line segments. This is an implementation of the equations presented at, among many other sources, https://paulbourke.net/geometry/pointlineplane/ . Bourke comments there, that "The equations apply to lines, if the intersection of line segments is required then it is only necessary to test if ua and ub lie between 0 and 1. Whichever one lies within that range then the corresponding line segment contains the intersection point. If both lie within the range of 0 to 1 then the intersection point is within both line segments."

Value

A matrix of the x and y coordinates of the intersection points. If no intersection exists, c(NA,NA) is returned. This is done so the return value is always a 2-element vector.

Author(s)

Carl Witthoft, carl@witthoft.com

See Also

polyInt

Extends rle to find and encode linear sequences.

Description

The function rle, or "run-length encoder," is a simple compression scheme which identifies sequences of repeating values in a vector. seqle extends this scheme by allowing the user to specify a sequence of values with a common "slope," or delta value, between adjacent elements. seqle with an increment of zero is the same as rle.

Usage

seqle(x, incr = 1L, prec = .Machine$double.eps^0.5)

Arguments

Details

Note: the returned value is assigned the class "rle". So far as I can tell, this class has only a print method, i.e. defining what is returned to the console when the user types the name of the returned object. Since the concept of "increment" has no reliable meaning when dealing with characters or char strings, when x is non-numeric the argument incr is ignored and the function reverts to base::rle.

Value

Note

The bulk of the code is taken directly from base::rle . Thanks to "flodel" on StackOverflow for suggesting code to handle floating-point increments.

Author(s)

Carl Witthoft, carl@witthoft.com

See Also

rle inverse.seqle

Examples

x<- c(2,2,2,3:8,8,8,4,4,4,5,5.5,6)
seqle(x,incr=0)
seqle(x,incr=1)
seqle(x,incr=1.5)

Returns a small sample of the specified data set.

Description

The user specifies both the number of elements to display and the number of elements at the start and end of the vector to ignore ('skip') when selecting elements. The results are displayed in a nice tabular form. There are options to set the value of N as well as the number of values to "skip" before selecting the values. short is similar to a combination of the unix head and tail functions.

Usage

short(x = seq(1, 20), numel = 4, skipel = 0, ynam = deparse(substitute(x)), dorows=FALSE)

Arguments

Details

If the argument x happens to be a list, short unlists everything, so the first numel values will be taken from the first list element, going on to the second element as needed, etc.

Value

Nothing is returned of interest. The function is called to provide what is printed directly to the console, which is a formatted table of the lead and tail values selected, with column labels identifying their location in the input vector object.

Author(s)

Carl Witthoft carl@witthoft.com

See Also

head , tail

Examples

foo<-matrix(runif(100),nrow=20)
short(foo)
short(foo,numel=6,skipel=10)
short(foo,numel=6,skipel=10,dorows=TRUE)

Execute simple zero-argument functions

Description

Execute simple zero-argument functions without having to type the "()" , and without having to go through the bother of makeActiveBinding. This code is provided primarily to allow the user to build his own set of command "shortcuts" by modifying the set of arguments to the switch function in the function body. The bulk of the code is copied from the excellent package sos . The name splatnd cannot be called directly, and doesn't even exist after being sourced. It serves to define a variety of operators ![your_string_here] . If the string after ! is not in the switch-list, the function defaults to the normal splat operator, i.e. NOT[your_string_here] .

Arguments

none

Details

The strings supported, with their associated functions, as delivered are: 'newdev' = dev.new(width=4.5, height= 4.5, restoreConsole=T), 'qapla' = cat('batlh tIn chav'), 'quitn' = quit('no'), 'quity' = quit('yes'), There's an obvious risk of undesired results should there exist an object in the environment with the same name as one of the items in the switch options. The workaround is to enclose the object name in parentheses. See the example.

Value

The returned value is the result of whatever function or operator was invoked.

Note

The bulk of the code is taken directly from the sos package.

Author(s)

Carl Witthoft, carl@witthoft.com

See Also

The R manuals on creating operators, findFn in the package sos , normally invoked as ???

Examples

# based on the default items in splatnd.R
qapla <- 1:5
!qapla
!(qapla)

Calculates the kurtosis of the input data set.

Description

Kurtosis is the next moment after skew (which is the moment after the variance). This function is provided primarily to support the function mystat . It uses the algorithm provided in the R package e1071

Usage

thekurt(x)

Arguments

Value

A single scalar, the kurtosis of the data provided.

Author(s)

Carl Witthoft, carl@witthoft.com

Calculates the skew of a dataset.

Description

This function is included primarily to support mystat . Skew is the next moment after the variance. The algorithm used here is taken from the R package e1071 .

Usage

theskew(x)

Arguments

Value

A single scalar, the skew of the dataset

Author(s)

Carl Witthoft, carl@witthoft.com