AndrewLJackson

---
title: "SIA biplots using ggplot2"
author: "Andrew L Jackson & Chris Harrod"
date: "`r Sys.Date()`"
output: html_notebook
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE, fig.width = 9, fig.height = 6)
```

AndrewLJackson

# examples of iteration estimates
# iteration 1, model 1
it1.m1<-c(0.007439839, 0.229489378, 0.035664652, 0.015154735, 0.712251396)
# iteration 2, model 1
it2.m1<-c(0.04197685, 0.22032073, 0.05718590, 0.03484489, 0.64567163)
# iteration 1, model 2
it1.m2<-c(0.09325976, 0.29549676, 0.15327778, 0.08599277, 0.37197294)
# iteration 2, model 2
it2.m2<- c(0.06429175, 0.02222554, 0.15422684, 0.65528134, 0.10397452) 

AndrewLJackson

# About: Code to calculate JDI and produce barplot as per
# https://scientistseessquirrel.wordpress.com/2017/08/03/my-journal-life-list/
# Additional code from
# https://www.r-bloggers.com/yet-another-post-on-google-scholar-data-analysis/
# Author: Andrew Jackson 
# http://www.tcd.ie/Zoology/research/research/theoretical/andrewjackson.php
# Date: 04-Aug-2017

AndrewLJackson

theta <- seq(0, 2*pi, length = 360)

Z <- data.frame(x = cos(theta), y = sin(theta))

par(mfrow = c(2,2))

# points in order
plot(y~x, data = Z, type = "l", main = "Made in order", asp = 1)

AndrewLJackson

library(dplyr)

# use the iris dataset
dd <- iris

# The normal way to group or split is by directly referencing the var object
dd %>% group_by(Species)

dd %>% split(.$Species)

AndrewLJackson

library(tidyverse)

# create a list of length 10 with a small dataframe in each entry
df <- lapply(1:10,function(x){data.frame(y = 1:5)})

# convert to tbl
df_tbl<- bind_rows(df, .id = "id")

# pluck out the last entry for each "id" using slice()
last_entry <- df_tbl %>% group_by(id) %>% slice(n())

AndrewLJackson

# code based on https://en.wikipedia.org/wiki/Mahalanobis_distance
set.seed(1)

# define a point D
x <- matrix(rmnorm(1, c(2,2), matrix(c(1,0.1,0.1,1), 2, 2)), 1, 2) # vector of locations in 2 dimensional space
x <- t(x) # transpose


# Define a multivariate normal distribution for comparison
mu <- matrix(c(0, 0), 1, 2) # vector of bivariate means

AndrewLJackson

strength = [82 86 79 83 84 85 86 87 74 82 ...
            78 75 76 77 79 79 77 78 82 79];
alloy = {'st','st','st','st','st','st','st','st',...
         'al1','al1','al1','al1','al1','al1',...
         'al2','al2','al2','al2','al2','al2'};

% NB also works on two-way anovas using anova2()
[p, tbl, stats] = anova1(strength,alloy,'off')

% this line will compute pairwise comparisons, at the usual 

AndrewLJackson

require(plyr)
#require(siar)

t1 <- data.frame(SizeClass=c("Meiofauna","Smaller Macrofauna","Bigger Macrofauna"),
           num=rpois(3,10),
           md13C=rnorm(3,-10,10),
           sd13C=rlnorm(3,1,1),
           md15N=rnorm(3,5,5) ,
           sd15N=rlnorm(3,1,1))
t1

AndrewLJackson

require(plyr)
require(siar)

t1 <- data.frame(SizeClass=c("Meiofauna","Smaller Macrofauna","Bigger Macrofauna"),
           num=rpois(3,10),
           md13C=rnorm(3,-10,10),
           sd13C=rlnorm(3,1,1),
           md15N=rnorm(3,5,5) ,
           sd15N=rlnorm(3,1,1))
t1