ggplot2.R

nSample <- 100
gender <- sample(c("male","female"),nSample,TRUE)
coffee <- pmax(0,
      round(rnorm(nSample,mean = ifelse(gender=="male",3,2))))
heartrate <- 80 + 20 * coffee + rnorm(nSample, 0, 20)
Data2 <- data.frame(gender=gender, coffee = coffee, 
                   heartrate = heartrate)
head(Data2)


library("ggplot2")
ggplot(Data2, aes(x = coffee, y = heartrate))


ggplot(Data2, aes(x = coffee, y = heartrate)) + 
  geom_point()


ggplot(Data2, aes(x = coffee, y = heartrate, colour = gender)) + 
  geom_point()


ggplot(Data2, aes(x = heartrate, y = coffee, colour = gender)) + 
  geom_point()


ggplot(Data2, aes(x=gender, y=heartrate, colour=coffee)) + 
  geom_point()


ggplot(Data2, 
  aes(x=gender, y=heartrate, colour=factor(coffee))) + 
  geom_point()


ggplot(Data2, 
  aes(x=gender, y=heartrate, colour=factor(coffee))) + 
  geom_boxplot()


ggplot(Data2, 
  aes(x = gender, y = heartrate, fill = factor(coffee))) + 
  geom_boxplot()


ggplot(Data2, 
  aes(x = gender, y = heartrate, fill = factor(coffee))) + 
  geom_boxplot() + xlab("Gender") + ylab("Heartrate") + 
  scale_fill_discrete("Coffee")


ggplot(Data2, 
  aes(x = gender, y = heartrate, fill = factor(coffee))) + 
  geom_boxplot() + xlab("Gender") + ylab("Heartrate") + 
  scale_fill_discrete("Coffee") + theme_bw()


ggplot(Data2, 
  aes(x = factor(coffee), y = heartrate, fill = gender)) + 
  geom_boxplot() + xlab("Coffee") + ylab("Heartrate") + 
  scale_fill_discrete("Gender") + theme_bw()


ggplot(Data2, 
  aes(x = coffee, y = heartrate, colour = gender)) + 
  geom_point() + xlab("Coffee") + ylab("Heartrate") + 
  scale_fill_discrete("Gender") + theme_bw() + 
  geom_smooth()


ggplot(Data2, 
  aes(x = coffee, y = heartrate, colour = gender)) + 
  geom_point() + xlab("Coffee") + ylab("Heartrate") + 
  scale_fill_discrete("Gender") + theme_bw() + 
  geom_smooth(method = "lm")


ggplot(Data2, 
  aes(x = coffee, y = heartrate, colour = gender)) + 
  geom_point() + xlab("Coffee") + ylab("Heartrate") + 
  scale_fill_discrete("Gender") + theme_bw() + 
  geom_smooth(method = "lm", se = FALSE)


ggplot(Data2, 
  aes(x = coffee, y = heartrate)) + 
  geom_point() + xlab("Coffee") + ylab("Heartrate") + 
  scale_fill_discrete("Gender") + theme_bw() + 
  geom_smooth(method = "lm", se = FALSE) +
  facet_grid(gender ~ .)


ggplot(Data2, aes(x = coffee, y = heartrate)) + 
  geom_point()


ggplot(Data2, aes(x = coffee, y = heartrate)) + 
  geom_point() + theme_bw()


ggplot(Data2, aes(x = coffee, y = heartrate)) + 
  geom_point() + theme_classic()


ggplot(Data2, aes(x = coffee, y = heartrate)) + geom_point() +
   theme(
    plot.background = element_rect(fill = "#E2E2E3", 
                                   colour = "#E2E2E3"),
    panel.background = element_rect(fill = "#E2E2E3"),
    axis.text = element_text(colour = "#E7A922"),
    plot.title = element_text(colour = "#552683", face = "bold", 
                              size = 18, vjust = 1),
    axis.title = element_text(colour = "#552683", face = "bold", 
                              size = 13),
    panel.grid.major.x = element_line(colour = "#E7A922"),
    panel.grid.minor.x = element_blank(),
    panel.grid.major.y = element_blank(),
    panel.grid.minor.y = element_blank(),
    strip.text = element_text(colour = "white"),
    strip.background = element_rect(fill = "#E7A922"),
    axis.ticks = element_line(colour = "#E7A922")
  )


p <- ggplot(Data2, aes(x = coffee, y = heartrate)) + geom_point() +
  theme(
      panel.background = element_rect(fill='transparent'),
      plot.background = element_rect(fill='transparent', color=NA),
      legend.background = element_rect(fill='transparent'),
      legend.box.background = element_rect(fill='transparent')
  )

ggsave("plot_trans.png", p,  width=10, height=6, dpi=300)

p <- ggplot(Data2, aes(x = coffee, y = heartrate)) + geom_point() +
  theme_classic() +
  theme(
      panel.background = element_rect(fill='transparent'),
      plot.background = element_rect(fill='transparent', color=NA),
      legend.background = element_rect(fill='transparent'),
      legend.box.background = element_rect(fill='transparent')
  )

ggsave("plot_trans_2.png", p,  width=4, height=2, dpi=300)

joins.R

# Long format data:
df_long <- data.frame(
  day = c("mon", "mon", "mon", "mon", "tue", "tue", 
          "tue", "tue", "wed", "wed", "wed", "wed"), 
  
  subject = c(1, 1, 2, 2, 1, 1, 2, 2, 1, 1, 2, 2), 
  
  variable = c("anxious", "sad", "anxious", "sad", 
               "anxious", "sad", "anxious", "sad", 
               "anxious", "sad", "anxious", "sad"), 
  
  value = c(1, 5, 1, 1, 2, 4, 2, 2, 1, 4, 1, 5)
)

# Subject data:
subject_data <- data.frame(
  id = c(1,2,3),
  name = c("Peter","Anne","Richard"),
  gender = c("male","female","male"),
  age = c(28,22,38)
)

subject_data

library("dplyr")
df_long %>% left_join(subject_data, by = c("subject" = "id"))

# Day data:
day_data <- data.frame(
  day = c("mon","tue","thu","fri"),
  temp = c(25,24,28,29)
)

day_data

df_long %>% left_join(day_data, by = "day")

df_long %>% inner_join(day_data, by = "day")

tidyr.R

df_long <- data.frame(
  day = c("mon", "mon", "mon", "mon", "tue", "tue", 
          "tue", "tue", "wed", "wed", "wed", "wed"), 
  
  subject = c(1, 1, 2, 2, 1, 1, 2, 2, 1, 1, 2, 2), 
  
  variable = c("anxious", "sad", "anxious", "sad", 
               "anxious", "sad", "anxious", "sad", 
               "anxious", "sad", "anxious", "sad"), 
  
  value = c(1, 5, 1, 1, 2, 4, 2, 2, 1, 4, 1, 5)
)

print(df_long)

library("ggplot2")
ggplot(df_long, aes(x = day, y = value, colour = factor(subject))) + 
  geom_point(size = 3) +
  facet_grid( ~ variable) + 
  theme_bw() + 
  scale_colour_discrete("Subject") + 
  ylab("") + xlab("")

library("dplyr")
library("tidyr")

df_wide <- df_long %>% pivot_wider(
    id_cols = c(day,subject), # The columns that will represent the rows in the new data
    names_from = variable, # Where to take the variable names from?
    values_from = value # Where to take the variable values from?
)

print(df_wide)

df_very_wide <- df_wide %>% 
  
  # Make Anxious label shorter (for slide):
  select(day,subject,anx=anxious,sad) %>% 
  
  # Make the data wider:
  pivot_wider(
  id_cols = subject, 
  names_from = day,
  values_from = c(sad,anx))

print(df_very_wide)

df_long %>% 
  
  # Make Anxious label shorter (for slide):
  mutate(variable = substr(variable,1,3)) %>%
  
  # Make the data wider:
  pivot_wider(
  id_cols = subject, 
  names_from = c(day,variable),
  values_from = value)

df_very_wide %>% pivot_longer(sad_mon:anx_wed)


df_very_wide %>% pivot_longer(sad_mon:anx_wed) %>% 
  separate(name,c("variable","day"))


df_very_wide %>% 
  pivot_longer(sad_mon:anx_wed,
               names_to = c("variable","day"),
               names_pattern = "(.*)_(.*)"
  )


# First make a data for the sad variables:
df_wide_sad <- df_very_wide %>% 
  
  # Select only the sad varibles:
  select(subject,sad_mon:sad_wed) %>%
  
  # Make longer:
  pivot_longer(sad_mon:sad_wed, 
               names_to = "day", 
               values_to = "sad") %>%
  
  # Remove the first 4 letters from the day string:
  mutate(day = substr(day,5,7))


print(df_wide_sad)


# Next the same for the anxious data:
df_wide_anx <- df_very_wide %>% 
  
  # Select only the sad varibles:
  select(subject,anx_mon:anx_wed) %>%
  
  # Make longer:
  pivot_longer(anx_mon:anx_wed, 
               names_to = "day", 
               values_to = "anxious") %>%
  
  # Remove the first 4 letters from the day string:
  mutate(day = substr(day,5,7))


print(df_wide_anx)


df_wide_sad %>% left_join(df_wide_anx)


library("reshape2")
df_long %>% dcast(day + subject ~ variable)


df_wide %>% melt(id.vars = c("subject","day"), 
                 measure.vars = c("anxious","sad"))