Rでデータ解析　その１

Rでデータ解析を行う。

> flour <- c(3,-2,-1,8,1,-2)
> diet <- c(-4,1,-3,-5,-2,-8)
> total<-c(flour, diet)

# histogramを書く

> hist(total)

> hist(total, br=c(-8,-6, -4, -2,0,2,4,6,8), ylim =c(0,5))

# 密度推定曲線

> plot(density(total), xlim=c(-8,8), ylim =c(0,0.2))

> plot(density(flour), xlim=c(-8,8), ylim =c(0,0.2), lty=1)

> par(new=T)
> plot(density(diet), xlim=c(-8,8), ylim =c(0,0.2), lty=2)

> legend(4,0.2,c("flour", "diet"), lty=1:2, ncol=1)

# 箱ひげグラフ

> boxplot(flour, diet, names=c("flour", "diet"))

> boxplot(total, flour, diet, names=c("total", "flour", "diet"))

# 合計

> sum(total)

[1] -14

> quantile(total)

   0%   25%   50%   75%  100% 
-8.00 -3.25 -2.00  1.00  8.00 

# min, 1Q median, 3Q, max

> fivenum(total)

[1] -8.0 -3.5 -2.0  1.0  8.0

# 平方和

> sum((total-mean(total))^2)

[1] 185.6667

> total

[1]  3 -2 -1  8  1 -2 -4  1 -3 -5 -2 -8

＃計算

> 3-2-1+8+1-2-4+1-3-5-2-8
[1] -14

# 標本分散

> variance <- function(x)var(x)*(length(x)-1)/length(x)
> variance(total)

[1] 15.47222

# 不偏分散

> var(total)

[1] 16.87879

# 標準誤差

> sd(total)

[1] 4.10838

> sqrt(var(total))

[1] 4.10838

# 不偏共分散行列

> var(flour, diet)

[1] -2.5

# 相関係数

> cor(flour, diet)

[1] -0.2142184