First, business things:

homework style & future grade improvement opportunity

final grade extra credit: 10 points for a report; 5 points for a presentation in class of the report
the report needs to be formatted specifically
good examples of a report: http://rpubs.com/gregmaghakian/438803
clean, minimal
all meat: every table/figure has a comment and a reason for inclusion, explain why you are including it
every word, every sentence should matter
other good example: https://rpubs.com/maypowerss/hw8
long intro, but very clean report overall
again, each table and figure makes sense and each inclusion matters for the overall narrative/report

ggplot2 lecture

read up on the first part of the slides, lots of good info & overview

napoleon 1812 winter retreat from moscow

war & peace figure: army invasion and retreat from moscow (pink is invasion; brown is retreat) the thickness of the lines represent size of army; also includes calendar time & temperature in the bottom section
“the best statistical figure ever created” & today we are going to try to replicate this graph using R

replicate War and Peace graphic using R

primary tool is ggplot2, plus a few other packages -HistData package install! this package includes lots of interesting datasets, including the map of London cholera outbreak

library(HistData)
troops <- Minard.troops
str(troops)

## 'data.frame':    51 obs. of  5 variables:
##  $ long     : num  24 24.5 25.5 26 27 28 28.5 29 30 30.3 ...
##  $ lat      : num  54.9 55 54.5 54.7 54.8 54.9 55 55.1 55.2 55.3 ...
##  $ survivors: int  340000 340000 340000 320000 300000 280000 240000 210000 180000 175000 ...
##  $ direction: Factor w/ 2 levels "A","R": 1 1 1 1 1 1 1 1 1 1 ...
##  $ group    : int  1 1 1 1 1 1 1 1 1 1 ...

#get the path of the troops
library(ggplot2)

## Warning: package 'ggplot2' was built under R version 3.4.4

ggplot(troops, aes(x = long, y = lat, group = group)) + geom_path()

#add color and some line thickness
ggplot(troops, aes(x = long, y = lat, group = group, color = direction, size = survivors)) + geom_path()

#slightly more polished by rounding the segments of the bars
ggplot(troops, aes(x = long, y = lat, group = group, color = direction, size = survivors)) + geom_path(lineend = "round")

#highlight the size of the change in size of the troops
ggplot(troops, aes(x = long, y = lat, group = group, color = direction, size = survivors)) + geom_path(lineend = "round") +
scale_size(range=c(.5, 15))

#this is an example range, but if you dont like the size of the segments, you can change the range values to tweak the appearance

#change the color to match that original figure
ggplot(troops, aes(x = long, y = lat, group = group, color = direction, size = survivors)) + 
  geom_path(lineend = "round") +
  scale_size(range=c(.5, 15)) + scale_colour_manual(values=c("#DFC17E", "#252523")) +
  labs(x = NULL, y = NULL) +
  guides(color = FALSE, size = FALSE)

#you can use r brewer to find hex codes, or use your favorite site

# adding city names
# we need this second dataset to get this info
# all the datasets we are using share latitude and longitude, which is how we can anchor them into the same graph
cities <- Minard.cities
str(cities)

## 'data.frame':    20 obs. of  3 variables:
##  $ long: num  24 25.3 26.4 26.8 27.7 27.6 28.5 28.7 29.2 30.2 ...
##  $ lat : num  55 54.7 54.4 54.3 55.2 53.9 54.3 55.5 54.4 55.3 ...
##  $ city: Factor w/ 20 levels "Bobr","Chjat",..: 5 18 15 9 4 7 16 13 1 19 ...

#add the city name data to the graph
ggplot() +
  geom_path(data = troops, aes(x = long, y = lat, group = group, color = direction, size = survivors), lineend = "round") +
  geom_point(data = cities, aes(x = long, y = lat)) +
  geom_text(data = cities, aes(x = long, y = lat, label = city), vjust = 1.5) +
  scale_size(range = c(0.5, 15)) + 
  scale_colour_manual(values = c("#DFC17E", "#252523")) +
  labs(x = NULL, y = NULL) + 
  guides(color = FALSE, size = FALSE)

#introduce first extra package
#the whole point of this package is to make it prettier
library(ggrepel)

## Warning: package 'ggrepel' was built under R version 3.4.4

#we can use this for some further refinements to change the font styles
ggplot() +
  geom_path(data = troops, aes(x = long, y = lat, group = group, 
                               color = direction, size = survivors),
            lineend = "round") +
  geom_point(data = cities, aes(x = long, y = lat),
             color = "#DC5B44") +
  geom_text_repel(data = cities, aes(x = long, y = lat, label = city),
                  color = "#DC5B44", family = "sans") +
  scale_size(range = c(0.5, 15)) + 
  scale_colour_manual(values = c("#DFC17E", "#252523")) +
  labs(x = NULL, y = NULL) + 
  guides(color = FALSE, size = FALSE)

# even better!
march.1812.plot.simple <- ggplot() +
  geom_path(data = troops, aes(x = long, y = lat, group = group, 
                               color = direction, size = survivors),
            lineend = "round") +
  geom_point(data = cities, aes(x = long, y = lat),
             color = "#DC5B44") +
  geom_text_repel(data = cities, aes(x = long, y = lat, label = city),
                  color = "#DC5B44", family = "sans") +
  scale_size(range = c(0.5, 15)) + 
  scale_colour_manual(values = c("#DFC17E", "#252523")) +
  labs(x = NULL, y = NULL) +
  guides(color = FALSE, size = FALSE) +
  theme_void()

march.1812.plot.simple

# add third dataset of temperature and time
library(dplyr)

## Warning: package 'dplyr' was built under R version 3.4.4

## 
## Attaching package: 'dplyr'

## The following objects are masked from 'package:stats':
## 
##     filter, lag

## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union

temps <- Minard.temp
temps.nice <- temps %>%
  mutate(nice.label = paste0(temp, "°, ", date))

## Warning: package 'bindrcpp' was built under R version 3.4.4

str(temps.nice)

## 'data.frame':    9 obs. of  5 variables:
##  $ long      : num  37.6 36 33.2 32 29.2 28.5 27.2 26.7 25.3
##  $ temp      : int  0 0 -9 -21 -11 -20 -24 -30 -26
##  $ days      : int  6 6 16 5 10 4 3 5 1
##  $ date      : Factor w/ 8 levels "Dec01","Dec06",..: 7 8 4 5 NA 6 1 2 3
##  $ nice.label: chr  "0°, Oct18" "0°, Oct24" "-9°, Nov09" "-21°, Nov14" ...

#put it all together
#use new data based on Minard.temp
temps.1812.plot <- ggplot(data = temps.nice, aes(x = long, y = temp)) +
  geom_line() +
  geom_label(aes(label = nice.label),
            family = "sans", size = 2.5) + 
  labs(x = NULL, y = "° Celsius") +
  scale_x_continuous(limits = ggplot_build(march.1812.plot.simple)$layout$panel_ranges[[1]]$x.range) +
  scale_y_continuous(position = "right") +
  coord_cartesian(ylim = c(-35, 5)) +  # Add some space above/below
  theme_bw(base_family = "sans") +
  theme(panel.grid.major.x = element_blank(),
        panel.grid.minor.x = element_blank(),
        panel.grid.minor.y = element_blank(),
        axis.text.x = element_blank(), axis.ticks = element_blank(),
        panel.border = element_blank())

temps.1812.plot

#combine the two plots together in a single graphic using the *gridExtra* package
# rbind is how you bind two figures; 
library(gridExtra)

## 
## Attaching package: 'gridExtra'

## The following object is masked from 'package:dplyr':
## 
##     combine

# Combine the two plots
both.1812.plot.simple <- gtable_rbind(ggplotGrob(march.1812.plot.simple),
                               ggplotGrob(temps.1812.plot))
#this is the meat; everything else below is to make it pretty

# Adjust panels
panels <- both.1812.plot.simple$layout$t[grep("panel", both.1812.plot.simple$layout$name)]

# Because this plot doesn't use coord_equal, since it's not a map, we can use whatever relative numbers we want, like a 3:1 ratio
both.1812.plot.simple$heights[panels] <- unit(c(3, 1), "null")

grid::grid.newpage()
grid::grid.draw(both.1812.plot.simple)

CREDIT

The code for replicating Minard’s 1812 plot was adapted from Andrew Heiss: https://github.com/andrewheiss/fancy-minard

what else can we do to improve the graphic?

add terrain maps to the background
add the scale of distance
connect the temperature to the points in troop movement with lines
add a caption
change the temperature measurement scale from celsius to fahrenheit
…anything else you want to do to change the graphic, you can do using ggplot2 and its family

lecture9_nov12-2018

First, business things:

homework style & future grade improvement opportunity

ggplot2 lecture

napoleon 1812 winter retreat from moscow

replicate War and Peace graphic using R

CREDIT

what else can we do to improve the graphic?