【R】二元配置分散分析
Rで統計解析するための備忘録的なやつです。
使用するデータ
このサイトの例題を使ってみます。
まずはデータフレームを作ります。
Soil <- factor(c("A","A","A","A","A","A","A","A","A","A","A","A",
"B","B","B","B","B","B","B","B","B","B","B","B"))
Fertilizer <- factor(c("100","100","100","200","200","200","300","300","300","400","400","400",
"100","100","100","200","200","200","300","300","300","400","400","400"))
Yield <- c(14.5, 15.1, 14.1, 16.5, 16.1, 15, 17.8, 19, 15.2, 18.1, 20.2, 17.2,
16.2, 15.3, 17.5, 18.6, 16.9, 18.6, 21.7, 20.5, 19.4, 23.6, 24.9, 25.5)
df <- data.frame(soil=Soil, fertilizer=Fertilizer, yield=Yield)
二元配置分散分析
aov()関数を使います。
aov()関数の書式は、aov(目的変数 ~ 説明変数, data = データフレーム名)です。
目的変数以外のデータフレームに含まれる残りの変数をすべて説明変数とするのであれば、
目的変数 ~.と表記することもできます。
繰り返しの無いデータの二元配置分散分析では、観測値に対する因子Aおよび因子Bの主効果の優位性を調べることができます。
一方、繰り返しのあるデータの二元配置分散分析では、因子Aおよび因子Bの各主効果の優位性に加えて、因子Aと因子Bとの交互作用についても調べることができます。
交互作用項を含めるときは因子A*因子Bを追加します。
今回の場合はsoil*fertilizerを加えます。
コードは以下のようになります。
summary()関数で分散分析表を出力します。
AOV <- aov(yield ~ soil + fertilizer +soil*fertilizer, data = df)
summary(AOV)
実行結果
> AOV <- aov(yield ~ soil + fertilizer +soil*fertilizer, data = df)
> summary(AOV)
Df Sum Sq Mean Sq F value Pr(>F)
soil 1 66.33 66.33 46.333 4.21e-06 ***
fertilizer 3 126.64 42.21 29.485 9.40e-07 ***
soil:fertilizer 3 17.79 5.93 4.142 0.0237 *
Residuals 16 22.91 1.43
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
最終的なスクリプト
# Rをきれいにする
rm(list = ls())
# データフレームの作成
Soil <- factor(c("A","A","A","A","A","A","A","A","A","A","A","A",
"B","B","B","B","B","B","B","B","B","B","B","B"))
Fertilizer <- factor(c("100","100","100","200","200","200","300","300","300","400","400","400",
"100","100","100","200","200","200","300","300","300","400","400","400"))
Yield <- c(14.5, 15.1, 14.1, 16.5, 16.1, 15, 17.8, 19, 15.2, 18.1, 20.2, 17.2,
16.2, 15.3, 17.5, 18.6, 16.9, 18.6, 21.7, 20.5, 19.4, 23.6, 24.9, 25.5)
df <- data.frame(soil=Soil, fertilizer=Fertilizer, yield=Yield)
# 二元配置分散分析（aov()関数）
AOV <- aov(yield ~ soil + fertilizer +soil*fertilizer, data = df)
summary(AOV)
参考：二元配置分散分析（lm()関数、anova()関数）
# Rをきれいにする
rm(list = ls())
# データフレームの作成
Soil <- factor(c("A","A","A","A","A","A","A","A","A","A","A","A",
"B","B","B","B","B","B","B","B","B","B","B","B"))
Fertilizer <- factor(c("100","100","100","200","200","200","300","300","300","400","400","400",
"100","100","100","200","200","200","300","300","300","400","400","400"))
Yield <- c(14.5, 15.1, 14.1, 16.5, 16.1, 15, 17.8, 19, 15.2, 18.1, 20.2, 17.2,
16.2, 15.3, 17.5, 18.6, 16.9, 18.6, 21.7, 20.5, 19.4, 23.6, 24.9, 25.5)
df <- data.frame(soil=Soil, fertilizer=Fertilizer, yield=Yield)
# 二元配置分散分析（aov()関数）
AOV <- aov(yield ~ soil + fertilizer +soil*fertilizer, data = df)
summary(AOV)
Rではlm()関数とanova()関数を組み合わせても分散分析を実行できます。
LM <- lm(yield ~ soil + fertilizer +soil*fertilizer, data = df)
anova(LM)
実行結果
> LM <- lm(yield ~ soil + fertilizer +soil*fertilizer, data = df)
> anova(LM)
Analysis of Variance Table
Response: yield
Df Sum Sq Mean Sq F value Pr(>F)
soil 1 66.334 66.334 46.3332 4.214e-06 ***
fertilizer 3 126.638 42.213 29.4850 9.403e-07 ***
soil:fertilizer 3 17.791 5.930 4.1423 0.02373 *
Residuals 16 22.907 1.432
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Author And Source
この問題について(【R】二元配置分散分析), 我々は、より多くの情報をここで見つけました https://qiita.com/insilicomab/items/9d32978a07eef5f8fdd9著者帰属：元の著者の情報は、元のURLに含まれています。著作権は原作者に属する。
Content is automatically searched and collected through network algorithms . If there is a violation . Please contact us . We will adjust (correct author information ,or delete content ) as soon as possible .