Thomas Lumley tslumley

An example of the non-transitivity of the Wilcoxon test, using Efron's non-transitive dice.

refron() generates numbers from a set of four non-transitive dice, so that a beats b beats c beats d beats a two-thirds of the time, with optional Gaussian smoothing.

example.R does Wilcoxon rank-sum tests to show that a > b > c > d > a

This is a model for NZ road deaths per billion km travelled, using data complied by Sam Warburton (@economissive on Twitter)

The model has an underlying smoothed trend which is a random walk with Gaussian increments, and a year-specific deviation that is also Gaussian. That is, it's a state space model with a random walk in the latent variable and a measurement model that's a Poisson-logNormal mixture to get overdispersed counts.

The variances of the random walk steps and the year-specific deviations have diffuse hyperpriors and the means are set up so that the mean of the exponential of the increment is 1 -- ie, so mu is genuinely the trend in means.

Data from the New Zealand Department of Corrections, provided to Mark Hanna under the Official Information Act

Original documents available at https://fyi.org.nz/request/3926-strip-searches-in-new-zealand-prisons#incoming-13386

Who gets interrupted on the Supreme Court

Data from Adam Feldman's Empirical SCOTUS blog: https://empiricalscotus.com/2016/05/26/interruptions/ https://empiricalscotus.com/2016/05/01/oral-arguments-2015/

as analysed at StatsChat: http://www.statschat.org.nz/2016/05/29/ima-let-you-finish/


	library(shiny)

	# Define UI for application that draws a histogram
	ui <- fluidPage(

	# Application title
	titlePanel("Bayesian Surprise"),

	# Sidebar with a slider input for number of bins

	# Based on https://gist.github.com/tjvananne/8b0e7df7dcad414e8e6d5bf3947439a9
	# Rewritten to work chunk by chunk, so I can read the 42B file with only 8GB memory

	# input .txt file, exports list of list of values and character vector of names (words)
	proc_pretrained_vec <- function(filename, chunksize=1000, guess_size=100000) {
	size<-guess_size
	here<-0

	# initialize space for values and the names of each word in vocab
	vals <- vector(mode = "list", length(size))

	library(tidyverse)
	library(leaflet)
	library(jsonlite)
	library(ratelimitr)
	library(htmlwidgets)
	library(htmltools)

	allroutes<-read_csv("~/Downloads/WLG-google-transit/routes.txt") %>% filter(route_type==3)

	download_route<-limit_rate(

	<!DOCTYPE html>
	<html lang="en" >

	<head>
	<meta charset="UTF-8">
	<title>Epiphany Clock</title>


	<link rel='stylesheet prefetch' href='http://maxcdn.bootstrapcdn.com/bootstrap/3.3.5/css/bootstrap.min.css'>

	pal<-function(mat) rgb(mat[,1],mat[,2],mat[,3],maxColorValue=255)


	showpal<-function(pal) {
	n<-length(pal)
	image(1:n, 1, as.matrix(1:n), col = pal,
	xlab = deparse(substitute(pal)), ylab = "", xaxt = "n", yaxt = "n",
	bty = "n")
	}


	gravity<-mks(c(3.7,8.9,9.8,3.7,23.1,9.0,8.7,11.0,0.6), m=1, kg=0, s=-2)
	density<-mks(c(5427,5243,5515,3933,1326,687,1270,1638,2390), m=-3,kg=1,s=0)

	speed.of.light<-fff(1.8026175e12,fur=1,fir=0,ftn=-1)

	setMethod("mean","mks",function(x,...){
	u<-getUnits(x)
	mks(mean(getValues(x)),u[1],u[2],u[3])
	})