Grammar of graphics in_class exercises2~5

library(dplyr)

library(ggplot2)
dta <- read.table("C:/Users/user/Desktop/langMathDutch.txt",header=TRUE)
head(dta)

dta1<- dta %>% mutate(class_size= cut(size,include.lowest = TRUE, breaks=quantile(size, probs=c(0, .33, .67, 1)), label=c("Small", "Medium", "Large"), ordered=T),IQV_f= cut(IQV, include.lowest = TRUE,breaks=quantile(IQV, probs=c(0, .33, .67, 1)), label=c("Low", "Middle", "High"), ordered=T))

table(dta1$groups)

library(ggplot2)
p0 <- ggplot(data=dta1, 
             mapping = aes(x=lang, 
                 y=arith)) +
  
  geom_point(shape = 23, fill = "black")+
  stat_smooth(formula= y ~ x,
              method='lm', 
              se=FALSE) +
  labs(x = "Language score",
       y = "Arithmetic score")+
   facet_wrap(.~class_size+IQV_f, labeller=labeller(.multi_line=F))
p0

dta <- datasets::USPersonalExpenditure
head(dta)

library(reshape2)
dta1 <- melt(dta)
names(dta1) <- c("category", "year", "expenditure")
dta1$expenditure <- log10(dta1$expenditure)
library(dplyr)
dta1 <- dta1 %>% mutate(excess = expenditure - mean(expenditure))
library(ggplot2)
qplot(excess, category, data = dta1) +
  geom_segment(aes(xend = 0, yend = category)) +
  geom_vline(xintercept = 0, colour = "grey50") +
  facet_wrap(~ year, nrow = 1)

dta <- WWGbook::autism
head(dta)

library(dplyr)
dta <- na.omit(dta)
dta <- dta %>% mutate(group = cut(sicdegp, breaks=c(0, 1, 2, 3),labels = c("L", "M", "H")),age_centered = age - mean(age)) 

library(ggplot2)

p1 <- ggplot(dta, aes(age_centered, vsae)) +
   facet_wrap(. ~ group)+
  geom_point(alpha = 0.40) +
  geom_smooth(method = "lm",
              formula = "y ~ x")+
  geom_path(aes(group = childid), alpha = 0.4)+
  theme_bw()+
  scale_x_continuous(limits = c(-4, 7.5),
                     breaks = c(-2.5, 0.0, 2.5, 5.0)) +
labs(x = "Age (in years, centered)",
       y = "VSAE score")


pd <- position_dodge(.3)
p2 <- dta %>% mutate(age_2 = age - 2) %>% group_by(group,age_2) %>%
  summarize(vsae_mean = mean(vsae), 
            vsae_se = sd(vsae) / sqrt(n())) %>%
  ggplot() + 
  aes(age_2, vsae_mean, 
      group=group, 
      shape=group) +
  geom_errorbar(aes(ymin=vsae_mean -vsae_se,
                    ymax=vsae_mean +vsae_se),
                width=.2, size=.3, 
                position=pd) +
  geom_line(position=pd, 
            show.legend = T,
            aes(linetype=group)) +
  geom_point(position=pd, 
             size=rel(3),
             show.legend = T) +
  scale_shape_manual(values = c(1, 2, 16)) +
  labs(x="Age (in year -2)", y="VSAE score") +
   theme_bw()+
   theme(legend.position=c(.08, .8),
        legend.box.background = element_rect(colour = "black"),
        legend.key = element_rect(color = "black"),
        legend.key.size = unit(1, "cm"))
p1

p2

dta <- read.table("C:/Users/user/Desktop/diabetes_mell.csv", sep = ",", header = T)
head(dta)

library(ggalluvial)
dta_v3 <- data.frame(with(dta[, c("race", "gender", "diabetes", "BMI")], 
                        xtabs(~ race + gender + diabetes + BMI)))
dta_v3$race <- factor(dta_v3$race, 
                    levels(dta_v3$race)[c(2,3,1)])
dta_v3$gender <- factor(dta_v3$gender, 
                    levels(dta_v3$gender)[c(2,1)])
dta_v3$diabetes <- factor(dta_v3$diabetes, 
                    levels(dta_v3$diabetes)[c(2,1)])
ggplot(dta_v3, 
       aes(axis1=race,
           axis2=gender, 
           axis3=diabetes, 
           y=Freq)) +
  scale_x_discrete(limits=c("race", 
                            "gender", 
                            "diabetes"), 
                   expand=c(.1, .05)) +
  labs(x='', 
       y='No. individuals') +
  geom_alluvium(aes(fill=BMI)) +
  geom_stratum() + 
  geom_text(stat="stratum", 
            infer.label=TRUE) +
  scale_fill_manual(values=c('gray40','tan1'))+
  theme_minimal() +
  theme(legend.position = "bottom") +
  ggtitle("Diabetes in overall population in US 2009-2010", subtitle = "straitified by race, gender and diabetes mellitus")