friendly’s gists

friendly / term_info.R

Last active February 12, 2023 16:36

Extract information about two-way terms in a loglinear model

	#' Extract information about two-way terms in a loglinear model
	#'
	#' This function is designed for use with an association graph showing the pairwise
	#' dependencies among categorical variables in a loglinear model for frequencies
	#' fit using \code{\link[MASS]{loglm}} or \code{\link[stats]{glm}} in a poisson family.
	#' The goal is to show
	#' an association graph diagram with edges indicating the strength of associations
	#' between each pair of variables.
	#'
	#' The weight for two-way terms can be found using the deviance or LRT or AIC statistic

friendly / test-newcommand.html

Created August 15, 2022 18:23

Quarto: Test LaTeX newcommands in HTML doc

	<!DOCTYPE html>
	<html xmlns="http://www.w3.org/1999/xhtml" lang="en" xml:lang="en"><head>

	<meta charset="utf-8">
	<meta name="generator" content="quarto-1.0.37">

	<meta name="viewport" content="width=device-width, initial-scale=1.0, user-scalable=yes">


	<title>Quarto: Test LaTeX newcommands in HTML doc</title>

friendly / clark-lord-paradox.R

Created February 7, 2022 19:08

Demonstrate Lord's Paradox example, following Michael Clarke

	# Demonstrate Lord's Paradox example, following Michael Clarke
	# https://m-clark.github.io/docs/lord/index.html

	library(ggplot2)
	library(dplyr)

	#' ## Data example: Initial and final weight for boys & girls
	#'
	#' Make a plot showing the distribution of gain scores perpendicular to the line of slope = 1
	#'

friendly / phys-models.bib

Created November 28, 2020 17:49

Bibliography for physical analogs of statistical models

	@comment{Physical analogs for statistical models}

	@article{GainesPalphrey:1932,
	ISSN = {00368075, 10959203},
	URL = {http://www.jstor.org/stable/1658643},
	author = {W. L. Gaines and J. R. Palfrey},
	journal = {Science},
	number = {1977},
	pages = {472--474},
	publisher = {American Association for the Advancement of Science},

friendly / PartySupport-2019.R

Created November 1, 2019 19:43

Plots of Canadian 2019 election results

	# Stacked Barplot of Support for Political Parites by Province in 2019 Canadian Federal Election
	# written by Christopher D. Green
	# 22 October 2019
	# additions by MF 31 Oct 2019

	provs<-c("BC","AB","SK","MB","ON","QC","NB","NS","PE","NF") # names of the provinces
	pops<-c(5.0,4.3,1.2,1.4,14.4,8.4,0.8,1.0,0.2,0.5) # populations in millions

	con<-c(34.1,69.2,64.3,45.4,33.2,16.0,32.8,25.7,27.4,28.0) # % support for Conservatives
	lib<-c(26.1,13.7,11.6,26.3,41.5,34.2,37.6,41.3,43.6,44.7) # % support for Liberals

friendly / minard-plot.R

Last active April 7, 2022 11:30

Reproduce Minard's March on Moscow Graphic using ggplot2

	#' # Reproduce Minard's March on Moscow Graphic using ggplot2

	#' This gist extends the examples given in the HistData package. It explicitly
	#' shows the construction of the plot as layers, and adds the plot of temperature
	#' vs. longitude below the plot of troop strength.

	#' ## Load data and required packages
	data(Minard.troops, package="HistData")
	data(Minard.cities, package="HistData")
	data(Minard.temp, package="HistData")

friendly / plot_r.R

Created November 29, 2016 01:25

	# Taken from: http://janhove.github.io/teaching/2016/11/21/what-correlations-look-like
	# source: http://janhove.github.io/RCode/plot_r.R
	#
	# plot_r
	#
	# Given a correlation coefficient and a sample size,
	# this function draws 16 scatterplots
	# whose shapes differ greatly
	# but all of which are consistent with the correlation coefficient.
	#

friendly / wavelength_to_rgb.R

Created October 24, 2016 19:50

Convert wavelength of color to RGB

	#' Wavelength to RGB
	#'
	#' This function converts a given wavelength of light to an
	#' approximate RGB color value.
	#'
	#' @param wavelength A wavelength value, in nanometers, in the human visual range from 380 nm through 750 nm.
	#' These correspond to frequencies in the range from 789 THz through 400 THz.
	#' @param gamma The \eqn{\gamma} correction for a given display device. The linear RGB values will require
	#' gamma correction if the display device has nonlinear response.
	#' @return a color string, corresponding to the result of \code{\link[grDevices]{rgb}} on the

friendly / 3d-demo.R

Created September 30, 2016 14:58

3D demonstrations of linear transformations and matrix inverse

	#' ---
	#' title: "3D demonstrations of linear transformations and matrix inverse"
	#' author: "Michael Friendly"
	#' date: "30 Sep 2016"
	#' ---

	#' Start with a unit cube, representing the identity matrix. Show its transformation
	#' by a matrix $A$ as the corresponding transformation of the cube.
	#'
	#' This also illustrates the determinant, det(A), as the volume of the transformed

friendly / boxM.R

Created February 10, 2016 15:12

Box's M-test for Homogeneity of Covariance Matrices

	# Box's M-test for Homogeneity of Covariance Matrices

	boxM <-
	function(mod, ...) UseMethod("boxM")


	boxM.default <- function(Y, group)
	{
	dname <- deparse(substitute(Y))
	if (!inherits(Y, c("data.frame", "matrix")))

Michael Friendly friendly