SciViews: SciViews - Main package

Description

Functions to install SciViews additions to R, and more tools.

Details

The SciViews package provides various functions to install the SciViews::R dialect. It also provides additional utilites besides base, recommended and tidyverse.

Important functions

TODO...

Author(s)

Maintainer: Philippe Grosjean phgrosjean@sciviews.org

See Also

Useful links:

• https://github.com/SciViews/SciViews

• https://www.sciviews.org/SciViews-R/

• Report bugs at https://github.com/SciViews/SciViews/issues

Various color palettes.

Description

Create vectors of 'n' contiguous colors.

Usage

rwb_colors(n, alpha = 1, s = 0.9, v = 0.9)

rwb.colors(n, alpha = 1, s = 0.9, v = 0.9)

rwg_colors(n, alpha = 1, s = 0.9, v = 0.9)

rwg.colors(n, alpha = 1, s = 0.9, v = 0.9)

ryg_colors(n, alpha = 1, s = 0.9, v = 0.9)

ryg.colors(n, alpha = 1, s = 0.9, v = 0.9)

cwm_colors(n, alpha = 1, s = 0.9, v = 0.9)

cwm.colors(n, alpha = 1, s = 0.9, v = 0.9)

Arguments

Details

'cwm_colors(s = 0.5, v = 1)' gives very similar colors to 'cm.colors()'. 'ryg_colors()' is similar to 'rainbow(start = 0, end = 2/6)'. The 'xxx_colors()' (tidyverse name-compatible) and 'xxx.colors()“ (grDevices name-compatible) functions are synonyms.

See Also

[grDevices::cm.colors()], [grDevices::colorRampPalette()]

Examples

# Draw color wheels with various palettes
opar <- par(mfrow = c(2, 2))
pie(rep(1, 11), col = cwm.colors(11), main = "Cyan - white - magenta")
pie(rep(1, 11), col = rwb.colors(11), main = "Red - white - blue")
pie(rep(1, 11), col = rwg.colors(11), main = "Red - white - green")
pie(rep(1, 11), col = ryg.colors(11), main = "Red - yellow - green")
par(opar)

Correlation matrices.

Description

Compute the correlation matrix between two variables, or more (between all columns of a matrix or data frame).

Usage

correlation(x, ...)

Correlation(x, ...)

## S3 method for class 'formula'
correlation(formula, data = NULL, subset, na.action, ...)

## Default S3 method:
correlation(
  x,
  y = NULL,
  use = "everything",
  method = c("pearson", "kendall", "spearman"),
  ...
)

is.Correlation(x)

is.correlation(x)

as.Correlation(x)

as.correlation(x)

## S3 method for class 'Correlation'
print(x, digits = 3, cutoff = 0, ...)

## S3 method for class 'Correlation'
summary(
  object,
  cutpoints = c(0.3, 0.6, 0.8, 0.9, 0.95),
  symbols = c(" ", ".", ",", "+", "*", "B"),
  ...
)

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

## S3 method for class 'Correlation'
plot(
  x,
  y = NULL,
  outline = TRUE,
  cutpoints = c(0.3, 0.6, 0.8, 0.9, 0.95),
  palette = rwb.colors,
  col = NULL,
  numbers = TRUE,
  digits = 2,
  type = c("full", "lower", "upper"),
  diag = (type == "full"),
  cex.lab = par("cex.lab"),
  cex = 0.75 * par("cex"),
  ...
)

## S3 method for class 'Correlation'
lines(
  x,
  choices = 1L:2L,
  col = par("col"),
  lty = 2,
  ar.length = 0.1,
  pos = NULL,
  cex = par("cex"),
  labels = rownames(x),
  ...
)

Arguments

Value

‘Correlation()' and 'as.Correlation()“ create a ’Correlation' object, while 'is.Correlation()“ tests for it.

There are ‘print()' and 'summary()' methods for the ’Correlation' object that differ in the symbolic encoding of the correlations in 'summary()', using symnum()], which makes large correlation matrices more readable.

The method 'plot()' returns nothing, but it draws ellipses on a graph that represent the correlation matrix visually. This is essentially the [ellipse::plotcorr()] function from package **ellipse**, with slightly different default arguments and with default 'cutpoints' equivalent to those used in the 'summary()' method.

Author(s)

Philippe Grosjean <phgrosjean@sciviews.org>, wrapping code in package ellipse, function [ellipse::plotcorr()] for the 'plot.Correlation()' method.

See Also

[stats::cov()], [stats::cov2cor()], [stats::cov.wt()], [stats::symnum()], [ellipse::plotcorr()] and look at [SciViews::panel_cor()]

Examples

# This is a simple correlation coefficient
cor(rnorm(10), runif(10))
Correlation(rnorm(10), runif(10))

# 'Correlation' objects allow better inspection of the correlation matrices
# than the output of default R cor() function
(longley.cor <- Correlation(longley))
summary(longley.cor) # Synthetic view of the correlation matrix
plot(longley.cor)    # Graphical representation

# Use of the formula interface
(mtcars.cor <- Correlation(~ mpg + cyl + disp + hp, data = mtcars,
  method = "spearman", na.action = "na.omit"))

mtcars.cor2 <- Correlation(mtcars, method = "spearman")
print(mtcars.cor2, cutoff = 0.6)
summary(mtcars.cor2)
plot(mtcars.cor2, type = "lower")

mtcars.cor2["mpg", "cyl"] # Extract a correlation from the correlation matrix

Enumerate items in an object.

Description

'enum()' is creating a vector of integers to enumarate items in an object. It is particularly useful in the 'for(i in enum(object))' construct.

Usage

enum(x)

Arguments

Note

The pattern 'for(i in 1:length(object))' is often found, but it fails in case 'length(object) == 0'! 'enum()' is indeed a synonym of 'seq_along()', but the later one is less expressive in the context.

See Also

[seq_along()]

Examples

enum(letters)
enum(numeric(0))
# Compare with:
1:length(numeric(0))
enum(NULL)
letters5 <- letters[1:5]
for (i in enum(letters5)) cat("letter", i, "=", letters5[i], "\n")

Logarithms.

Description

To avoid confusion using the default 'log()' function, which is natural logarithm, but spells out like base 10 logarithm in the mind of some beginneRs, we define 'ln()' and 'ln1p()' as wrappers for 'log()“ with default 'base = exp(1)' argument and for 'log1p()', respectively. For similar reasons, 'lg()' is a wrapper of 'log10()' (there is no possible confusion here, but 'lg' is another common notation for base 10 logarithm). 'lg1p()' is a convenient way to use the optimized code to calculate the logarithm of x + 1, but returning the result in base 10 logarithm. 'E' is the Euler constant and is provided for convenience as 'exp(1)'. Finally 'lb()' is a synonym of 'log2()'.

Usage

ln(x)

ln1p()

lg()

lg1p(x)

E

lb()

Arguments

Format

An object of class numeric of length 1.

See Also

[log()]

Examples

ln(exp(3))              # Same as log(exp(3))
ln1p(c(0, 1, 10, 100))  # Wrapper for log1p()
lg(10^3)                # Same as log10(10^3)
lg1p(c(0, 1, 10, 100))  # log10(x + 1), but optimized for x << 1
E^4                     # Similar to exp(4), but different calculation!
lb(1:3)                 # Wrapper for log2()

Convenience functions for rows or columns manipulations.

Description

'nr()' and 'nc()' are synonyms of the ugly 'NROW()' or 'NCOL()' that still provide a result, even if 'dim' attribute of the object is not set, on the contrary to 'nrow()'or 'ncol()'. 'ROWS' and 'COLS' are constants that makes call to 'apply()' more expressive.

Usage

nr(x)

nc(x)

ROWS

COLS

Arguments

Format

An object of class numeric of length 1.

An object of class numeric of length 1.

See Also

[NROW()]

Examples

mm <- matrix(1:6, nrow = 3)
nr(mm)
nc(mm)
vv <- 1:6
nr(vv)
nc(vv)
# ROWS and COLS constants used with apply()
apply(mm, ROWS, mean) # Idem apply(mm, 1, mean)
apply(mm, COLS, mean) # Idem apply(mm, 2, mean)

More panel plots.

Description

Several panel plots that can be used with functions like [graphics::coplot()] and [graphics::pairs))].

Usage

panel_reg(
  x,
  y,
  col = par("col"),
  bg = par("bg"),
  pch = par("pch"),
  cex = par("cex"),
  lwd = par("lwd"),
  line.reg = lm,
  line.col = "red",
  line.lwd = lwd,
  untf = TRUE,
  ...
)

panel.reg(
  x,
  y,
  col = par("col"),
  bg = par("bg"),
  pch = par("pch"),
  cex = par("cex"),
  lwd = par("lwd"),
  line.reg = lm,
  line.col = "red",
  line.lwd = lwd,
  untf = TRUE,
  ...
)

panel_ellipse(
  x,
  y,
  col = par("col"),
  bg = par("bg"),
  pch = par("pch"),
  cex = par("cex"),
  el.level = 0.7,
  el.col = "cornsilk",
  el.border = "red",
  major = TRUE,
  ...
)

panel.ellipse(
  x,
  y,
  col = par("col"),
  bg = par("bg"),
  pch = par("pch"),
  cex = par("cex"),
  el.level = 0.7,
  el.col = "cornsilk",
  el.border = "red",
  major = TRUE,
  ...
)

panel_cor(
  x,
  y,
  use = "everything",
  method = c("pearson", "kendall", "spearman"),
  alternative = c("two.sided", "less", "greater"),
  digits = 2,
  prefix = "",
  cex = par("cex"),
  cor.cex = cex,
  stars.col = "red",
  ...
)

panel.cor(
  x,
  y,
  use = "everything",
  method = c("pearson", "kendall", "spearman"),
  alternative = c("two.sided", "less", "greater"),
  digits = 2,
  prefix = "",
  cex = par("cex"),
  cor.cex = cex,
  stars.col = "red",
  ...
)

panel_smooth(
  x,
  y,
  col = par("col"),
  bg = NA,
  pch = par("pch"),
  cex = 1,
  col.smooth = "red",
  span = 2/3,
  iter = 3,
  ...
)

Arguments

Details

Theses functions should be used outside of the diagonal in [graphics::pairs()], or with [graphics::coplot()], as they are bivariate plots.

Value

These functions return nothing and are used for their side effect of plotting in panels of composite plots.

Author(s)

Philippe Grosjean <phgrosjean@sciviews.org>, but code inspired from [graphics::panel.smooth()] in **graphics** and 'panel.car()' in package **car**.

See Also

[graphics::coplot()], [graphics::pairs()], [graphics::panel.smooth()], [stats::lm()], [ellipse::ellipse()], [stats::cor()] and [stats::cor.test()]

Examples

# Smooth lines in lower graphs and straight lines in upper graphs
pairs(trees, lower.panel = panel_smooth, upper.panel = panel_reg)
# Robust regression lines
library(MASS)  # For rlm()
pairs(trees, panel = panel_reg, diag.panel = panel_boxplot,
  reg.line = rlm, line.col = "blue", line.lwd = 2)
# A Double log graph
pairs(trees, lower.panel = panel_smooth, upper.panel = panel_reg, log = "xy")

# Graph suitables to explore correlations (take care there are potentially
# many simultaneous tests done here... So, you loose much power in the whole
# analysis... use it just as an indication!)
# Pearson's r
pairs(trees, lower.panel = panel_ellipse, upper.panel = panel_cor)
# Spearman's rho (ellipse and straight lines not suitable here!)
pairs(trees, lower.panel = panel_smooth, upper.panel = panel_cor,
  method = "spearman", span = 1)
# Several groups (visualize how bad it is to consider the whole set at once!)
pairs(iris[, -5], lower.panel = panel_smooth, upper.panel = panel_cor,
  method = "kendall", span = 1,
  col = c("red3", "blue3", "green3")[iris$Species])
# Now analyze correlation for one species only
pairs(iris[iris$Species == "virginica", -5], lower.panel = panel_ellipse,
  upper.panel = panel_cor)

# A coplot with custom panes
coplot(Petal.Length ~ Sepal.Length | Species, data = iris,
  panel = panel_ellipse)

More univariate panel plots.

Description

Several panel plots that can be used with [graphics::pairs()].

Usage

panel_boxplot(x, col = par("col"), box.col = "cornsilk", ...)

panel.boxplot(x, col = par("col"), box.col = "cornsilk", ...)

panel_density(
  x,
  adjust = 1,
  rug = TRUE,
  col = par("col"),
  lwd = par("lwd"),
  line.col = col,
  line.lwd = lwd,
  ...
)

panel.density(
  x,
  adjust = 1,
  rug = TRUE,
  col = par("col"),
  lwd = par("lwd"),
  line.col = col,
  line.lwd = lwd,
  ...
)

panel_hist(
  x,
  breaks = "Sturges",
  hist.col = "cornsilk",
  hist.border = NULL,
  hist.density = NULL,
  hist.angle = 45,
  ...
)

panel.hist(
  x,
  breaks = "Sturges",
  hist.col = "cornsilk",
  hist.border = NULL,
  hist.density = NULL,
  hist.angle = 45,
  ...
)

panel_qqnorm(
  x,
  pch = par("pch"),
  col = par("col"),
  bg = par("bg"),
  cex = par("cex"),
  lwd = par("lwd"),
  qq.pch = pch,
  qq.col = col,
  qq.bg = bg,
  qq.cex = cex,
  qqline.col = qq.col,
  qqline.lwd = lwd,
  ...
)

panel.qqnorm(
  x,
  pch = par("pch"),
  col = par("col"),
  bg = par("bg"),
  cex = par("cex"),
  lwd = par("lwd"),
  qq.pch = pch,
  qq.col = col,
  qq.bg = bg,
  qq.cex = cex,
  qqline.col = qq.col,
  qqline.lwd = lwd,
  ...
)

Arguments

Details

Panel functions [SciViews::panel_boxplot()], [SciViews::panel_density()], [SciViews::panel_hist()] and [SciViews::panel_qqnorm()] should be used only to plot univariate data on the diagonals of pair plots (or scatterplot matrix).

Value

These functions return nothing and are used for their side effect of plotting in panels of composite plots.

Author(s)

Philippe Grosjean <phgrosjean@sciviews.org>, but code inspired from 'spm()' in package **car**.

See Also

[graphics::pairs()], [graphics::boxplot()], [graphics::hist()], [stats::density()], [stats::qqnorm()]

Examples

# Example of scatterplot matrices with custom plots on the diagonal
# Boxplots
pairs(trees, panel = panel_smooth, diag.panel = panel_boxplot)
pairs(trees, diag.panel = panel_boxplot, box.col = "gray")
# Densities
pairs(trees, panel = panel_smooth, diag.panel = panel_density)
pairs(trees, diag.panel = panel_density, line.col = "red", adjust = 0.5)
# Histograms
pairs(trees, panel = panel_smooth, diag.panel = panel_hist)
pairs(trees, diag.panel = panel_hist, hist.col = "gray", breaks = "Scott")
# QQ-plots against Normal theoretical distribution
pairs(trees, panel = panel_smooth, diag.panel = panel_qqnorm)
pairs(trees, diag.panel = panel_qqnorm, qqline.col = 2, qq.cex = .5, qq.pch = 3)

Principal Components Analysis.

Description

Perform a principal components analysis on a matrix or data frame and return a 'pcomp' object.

Usage

pcomp(x, ...)

## S3 method for class 'formula'
pcomp(formula, data = NULL, subset, na.action, method = c("svd", "eigen"), ...)

## Default S3 method:
pcomp(
  x,
  method = c("svd", "eigen"),
  scores = TRUE,
  center = TRUE,
  scale = TRUE,
  tol = NULL,
  covmat = NULL,
  subset = rep(TRUE, nrow(as.matrix(x))),
  ...
)

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

## S3 method for class 'pcomp'
summary(object, loadings = TRUE, cutoff = 0.1, ...)

## S3 method for class 'summary.pcomp'
print(x, digits = 3, loadings = x$print.loadings, cutoff = x$cutoff, ...)

## S3 method for class 'pcomp'
plot(
  x,
  which = c("screeplot", "loadings", "correlations", "scores"),
  choices = 1L:2L,
  col = par("col"),
  bar.col = "gray",
  circle.col = "gray",
  ar.length = 0.1,
  pos = NULL,
  labels = NULL,
  cex = par("cex"),
  main = paste(deparse(substitute(x)), which, sep = " - "),
  xlab,
  ylab,
  ...
)

## S3 method for class 'pcomp'
screeplot(
  x,
  npcs = min(10, length(x$sdev)),
  type = c("barplot", "lines"),
  col = "cornsilk",
  main = deparse(substitute(x)),
  ...
)

## S3 method for class 'pcomp'
points(
  x,
  choices = 1L:2L,
  type = "p",
  pch = par("pch"),
  col = par("col"),
  bg = par("bg"),
  cex = par("cex"),
  ...
)

## S3 method for class 'pcomp'
lines(
  x,
  choices = 1L:2L,
  groups,
  type = c("p", "e"),
  col = par("col"),
  border = par("fg"),
  level = 0.9,
  ...
)

## S3 method for class 'pcomp'
text(
  x,
  choices = 1L:2L,
  labels = NULL,
  col = par("col"),
  cex = par("cex"),
  pos = NULL,
  ...
)

## S3 method for class 'pcomp'
biplot(x, choices = 1L:2L, scale = 1, pc.biplot = FALSE, ...)

## S3 method for class 'pcomp'
pairs(
  x,
  choices = 1L:3L,
  type = c("loadings", "correlations"),
  col = par("col"),
  circle.col = "gray",
  ar.col = par("col"),
  ar.length = 0.05,
  pos = NULL,
  ar.cex = par("cex"),
  cex = par("cex"),
  ...
)

## S3 method for class 'pcomp'
predict(object, newdata, dim = length(object$sdev), ...)

## S3 method for class 'pcomp'
correlation(x, newvars, dim = length(x$sdev), ...)

scores(x, ...)

## S3 method for class 'pcomp'
scores(x, labels = NULL, dim = length(x$sdev), ...)

Arguments

Details

'pcomp()' is a generic function with '"formula"' and '"default"' methods. It is essentially a wrapper around [stats::prcomp()] and [stats::princomp()] to provide a coherent interface and object for both methods.

A 'pcomp' object is created. It inherits from 'pca' (as in **labdsv** package, but not compatible with the 'pca' object of package **ade4**) and of 'princomp'.

For more information on calculation done, refer to [stats::prcomp()] for 'method = "svd"' or [stats::princomp()] for 'method = "eigen"'.

Value

A 'c("pcomp", "pca", "princomp")' object.

Note

The signs of the columns of the loadings and scores are arbitrary, and so may differ between functions for PCA, and even between different builds of R.

Author(s)

Philippe Grosjean <phgrosjean@sciviews.org>, but the core code is indeed in package **stats**.

See Also

[vectorplot()], [stats::prcomp()], [stats::princomp()], [stats::loadings()], [SciViews::Correlation()]

Examples

# We will analyze mtcars without the Mercedes data (rows 8:14)
data(mtcars)
cars.pca <- pcomp(~ mpg + cyl + disp + hp + drat + wt + qsec, data = mtcars,
  subset = -(8:14))
cars.pca
summary(cars.pca)
screeplot(cars.pca)

# Loadings are extracted and plotted like this
(cars.ldg <- loadings(cars.pca))
plot(cars.pca, which = "loadings") # Equivalent to vectorplot(cars.ldg)

# Similarly, correlations of variables with PCs are extracted and plotted
(cars.cor <- Correlation(cars.pca))
plot(cars.pca, which = "correlations") # Equivalent to vectorplot(cars.cor)
# One can add supplementary variables on this graph
lines(Correlation(cars.pca,
  newvars = mtcars[-(8:14), c("vs", "am", "gear", "carb")]))

# Plot the scores
plot(cars.pca, which = "scores", cex = 0.8) # Similar to plot(scores(x)[, 1:2])
# Add supplementary individuals to this plot (labels), also points() or lines()
text(predict(cars.pca, newdata = mtcars[8:14, ]), col = "gray", cex = 0.8)

# Pairs plot for 3 PCs
iris.pca <- pcomp(iris[, -5])
pairs(iris.pca, col = (2:4)[iris$Species])

Timing of R expressions.

Description

Similar to ‘system.time()' but returns a more convenient ’difftime' object.

Usage

timing(expr, gc.first = TRUE)

Arguments

See Also

[system.time()]

Examples

test <- timing(Sys.sleep(0.5))
test
attr(test, "details")

Plot vectors inside a unit circle (PCA loadings or correlations plots).

Description

Plots vectors with 0 < norms < 1 inside a circle. These plots are mainly designed to represent variables in principal components space for PCAs.

Usage

vectorplot(x, ...)

## Default S3 method:
vectorplot(
  x,
  y,
  col = par("col"),
  circle.col = "gray",
  ar.length = 0.1,
  pos = NULL,
  cex = par("cex"),
  labels = NULL,
  ...
)

## S3 method for class 'loadings'
vectorplot(
  x,
  choices = 1L:2L,
  col = par("col"),
  circle.col = "gray",
  ar.length = 0.1,
  pos = NULL,
  cex = par("cex"),
  labels = rownames(x),
  main = deparse(substitute(x)),
  ...
)

## S3 method for class 'Correlation'
vectorplot(
  x,
  choices = 1L:2L,
  col = par("col"),
  circle.col = "gray",
  ar.length = 0.1,
  pos = NULL,
  cex = par("cex"),
  labels = rownames(x),
  main = deparse(substitute(x)),
  ...
)

Arguments

Value

The object 'x' is returned invisibly. These functions are called for their side-effect of drawing a vector plot.

See Also

[SciViews::pcomp()], [stats::loadings()], [SciViews::Correlation()]

Examples

# Create a PCA and plot loadings and correlations
iris.pca <- pcomp(iris[, -5])
vectorplot(loadings(iris.pca))
vectorplot(Correlation(iris.pca))
# Note: on screen devices, change aspect ratio of the graph by resizing
# the window to reveal cropped labels...