pratica.R

# O dataset utilizado será o Iris, disponibilizado pelo R

?iris

class(iris)

head(iris)

# ESTATÍSTICA DESCRITIVA

# Somatório

sum(iris$Sepal.Length)

# Produtório

prod(iris$Sepal.Length)

# Média arítmética

mean(iris$Sepal.Length)

sum(iris$Sepal.Length)/length(iris$Sepal.Length)

mean(iris$Sepal.Length[iris$Species == "setosa"])

# Média Ponderada

seto <- mean(iris$Sepal.Length[iris$Species == "setosa"])

vers <- mean(iris$Sepal.Length[iris$Species == "versicolor"])

virg <-mean(iris$Sepal.Length[iris$Species == "virginica"])

(seto * 5 + vers * 3 + virg * 2)/10

# Mediana

median(iris$Sepal.Length)

# Moda

getmode <- function(v) {uniqv <- unique(v) uniqv[which.max(tabulate(match(v, uniqv)))]}

getmode(iris$Sepal.Width)

table(iris$Sepal.Width)

# Variância de Desvio Padrão

var(iris$Sepal.Length)

sd(iris$Sepal.Length)

sqrt(var(iris$Sepal.Length))

# Outras métricas

min(iris$Sepal.Length)

max(iris$Sepal.Length)

range(iris$Sepal.Length)

quantile(iris$Sepal.Length)

abs(iris$Sepal.Length)

sample(iris$Sepal.Length, 10)

summary(iris$Sepal.Length)

table(iris$Sepal.Length)

# Correlação

cor(iris$Sepal.Length,iris$Petal.Length)

cor(iris$Petal.Length,iris$Petal.Width)

cor(iris$Petal.Width,iris$Sepal.Width)

# GRÁFICOS

# line plots

plot(table(iris$Petal.Width), col='blue', type='o')

# Scatter plot

scatter.smooth(iris$Petal.Length~iris$Petal.Width,col=heat.colors(length(iris$Petal.Length)),pch=16)

#Bar plot
barplot(table(iris$Sepal.Length),col=cm.colors(length(table(iris$Sepal.Length))))

# Pie plot

pie(head(table(iris$Petal.Length),10))

# Hisograma

hist(iris$Sepal.Length)

# Density plot

plot(density(iris$Sepal.Length))

# Boxplot

boxplot(iris)

summary(iris)

# Heatmap
heatmap(t(head(iris[1:4], 30)))

##
## ggplot2
##

library(ggplot2)

#scatter plot simples - camadas 1!

ggplot(data=iris, aes(x = Sepal.Length, y= Petal.Length))

#scatter plot simples - camadas 2!
ggplot(data=iris, aes(x = Sepal.Length, y= Petal.Length)) + geom_point()

#scatter plot melhorado 1

ggplot(data=iris, aes(x = iris$Sepal.Length, y= iris$Petal.Length))+ geom_point(aes(col=iris$Species))

#scatter plot melhorado 2

ggplot(data=iris, aes(x = iris$Sepal.Length, y= iris$Petal.Length))+ geom_point(aes(col=iris$Species, shape=iris$Species))

#scatter plot melhorado 3
ggplot(data=iris, aes(x = iris$Sepal.Length, y= iris$Petal.Length))+ geom_point(aes(col=iris$Species, shape=iris$Species)) + geom_smooth()


#scatter plot separado

ggplot(data=iris, aes(x = iris$Sepal.Length, y= iris$Petal.Length))+ geom_point(aes(col=iris$Species, shape=iris$Species)) + facet_grid(. ~ Species)


#boxplot

ggplot(data=iris, aes(x=iris$Species, y=iris$Sepal.Length)) + geom_boxplot(aes(fill=iris$Species))


#Violin plot

ggplot(data=iris, aes(x=iris$Species, y=iris$Sepal.Length)) + geom_violin(aes(fill=iris$Species))

#Histograma

ggplot(data=iris, aes(x=iris$Sepal.Width)) + geom_histogram(binwidth=0.2, color="black", aes(fill=Species))

# Densidade

ggplot(data=iris, aes(x=Sepal.Width, fill=Species))+ geom_density(stat="density", alpha=(0.2))

# heatmap

library(reshape2)

dat <- iris[,1:4]
cor <- melt(cor(dat, use="p"))
head(cor)

ggplot(data=cor, aes(x=Var1, y=Var2, fill=value)) + scale_fill_gradient2(limits=c(-1, 1))

#Adicionando rótulos
ggplot(data=cor, aes(x=Var1, y=Var2, fill=value)) + scale_fill_gradient2(limits=c(-1, 1)) + geom_tile()+ geom_text(aes(label=round(value,3))