Learn R Programming

⚠️There's a newer version (1.17.1) of this package.Take me there.

shipunov (version 1.6)

Miscellaneous Functions from Alexey Shipunov

Description

A collection of functions for data manipulation, plotting and statistical computing, to use separately or with the book "Visual Statistics. Use R!": Shipunov (2019) . Most useful functions are probably Bclust(), Jclust() and BootA() which bootstrap hierarchical clustering; Recode...() which multiple recode in a fast, flexible and simple way; Misclass() which outputs confusion matrix even if classes are not concerted; Overlap() which calculates overlaps of convex hulls from any projection; and Pleiad() which is fast and flexible correlogram. In fact, there are much more useful functions, please see documentation.

Copy Link

Version

Install

install.packages('shipunov')

Monthly Downloads

286

Version

1.6

License

GPL (>= 2)

Maintainer

Alexey Shipunov

Last Published

March 4th, 2020

Functions in shipunov (1.6)

Adjusted Rand index

Samples along the class labels

Adds correlation arrows to the scatterplot

Convert diagnostic keys and classification lists

Calculates correlation and converts results into the named long vector

Bootstrap clustering

Bootstrap with kNN

Cosine distance

Class projection

Examples of colors

Classification grid and decision boundaries

Average coefficients of determination for each variable

Compare checklists

Clustering to matrix

Dotchart-like plot sfor every scaled variable grouped by factor

Improved dotcharts

Textual file system browser

Correlation matrix with p-values

Check normality

Matrix Representation of Hierarchical Clustering

Find local maxima

Rarefaction curves

Fill data values downstream, like in spreadsheets

MDS: explained variance (surrogate)

Simple bootstrap and jackknife clustering

Polygons' overlap

Examples of fonts

Bootstrapped hclust

Plant phyllotaxis

Calculates the best overlap

Converts 'rpart' object into Newick tree

Coefficients of variation

Rresults shell script

System date, time plus easy save history

Say ``no'' to dynamite plots!

Roots phylogenetic trees even if outgroup is not monophyletic

Generates datasets for clustering

Histogram with overlaid curve

Polygons' squares

Haltica flea beetles

Educated distances for semi-supervised clustering

Calculates multiple correlation matrices (via 'factor1') and stacks them together

Stacks correlation matrix

Chaetocnema flea beetles

Classification lists

Groups' confidence ellipses

Boxplot explanation

Textual plot of missing data

Examples of plot types

Counts matches between two hierarchical clusterings

Finds all duplicates

Plots dotchart with best distance method, use multiple non-base distances

Read 'NTSYSpc' files

Diagnostic keys

Basic multiple recoding

'str()' enchanced for data frames

Changes the appearance of cluster dendrogram

Plot bootstrap values on 'hclust' plot

Misclassification (confusion) table

Plot bootstrap values

Convert table to data frame saving structure

Bootstrap with 'randomForest()'

Plot phylogenetic tree with clades collapsed

Distance matrix based kNN classification

Removes duplicated data values downstream

Grouped boxplots

Correlation circles (correlation pleiads)

Select rows from data frame

Robust rank order test

Number of cases in each location reflected in the point size

Examples of line types

Separate terminal pager for Linux

Example of plot margins

pairwise.Rro.test

Robust rank order test post hoc derivative

Coefficient of divergence

Interpreter for effect sizes

Plots dotchart with best base distance method

PlotBest.hclust

Plots dotchart with best clustering method

Imitation (!) of the modern 'R' logo

Read 'FASTA' files

pairwise.Table2.test

Pairwise Chi-squared or Fisher test for 2-dimensional tables

Binarize (make dummy variables)

Effect sizes of association between categorical variables

Write 'FASTA' files

Pairwise table of effects with magnitudes

Insert content to Linux X11 clipboard

Aggregates by one vector and uses it for row names

Adds confidence bands to the simple linear model plots