ANOVA

Librerias

library(readxl)
library(dplyr)

## 
## Attaching package: 'dplyr'

## The following objects are masked from 'package:stats':
## 
##     filter, lag

## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union

library(MASS)

## 
## Attaching package: 'MASS'

## The following object is masked from 'package:dplyr':
## 
##     select

library(ggplot2)

Importar datos

data<-read_excel('C:/Users/rusoc/OneDrive/Escritorio/TEC/Mineria de datos/Diet.xlsx')

str(data)

## tibble [78 × 4] (S3: tbl_df/tbl/data.frame)
##  $ gender      : chr [1:78] "M" "M" "F" "F" ...
##  $ Diet        : chr [1:78] "B" "B" "A" "A" ...
##  $ weight      : num [1:78] 60 103 58 60 64 64 65 66 67 69 ...
##  $ weight6weeks: num [1:78] 60 103 54.2 54 63.3 61.1 62.2 64 65 60.5 ...

table(data$Diet)

## 
##  A  B  C 
## 24 27 27

Vamos a crear una nueva columna para ver la perdida de peso que hubo en las 6 semanas.

data2 <- data%>% mutate(perdida_peso= data$weight-data$weight6weeks)
data2

## # A tibble: 78 × 5
##    gender Diet  weight weight6weeks perdida_peso
##    <chr>  <chr>  <dbl>        <dbl>        <dbl>
##  1 M      B         60         60          0    
##  2 M      B        103        103          0    
##  3 F      A         58         54.2        3.80 
##  4 F      A         60         54          6    
##  5 F      A         64         63.3        0.700
##  6 F      A         64         61.1        2.9  
##  7 F      A         65         62.2        2.80 
##  8 F      A         66         64          2    
##  9 F      A         67         65          2    
## 10 F      A         69         60.5        8.5  
## # ℹ 68 more rows

aggregate(perdida_peso~Diet, data = data2, FUN = mean)

##   Diet perdida_peso
## 1    A     3.300000
## 2    B     3.025926
## 3    C     5.148148

Creamos boxplox

ggplot(data = data2, aes(x=Diet, y=perdida_peso, color=Diet)) + geom_boxplot() + theme_bw()

anova = aov(data2$perdida_peso ~ data2$Diet)
summary(anova)

##             Df Sum Sq Mean Sq F value  Pr(>F)   
## data2$Diet   2   71.1   35.55   6.197 0.00323 **
## Residuals   75  430.2    5.74                   
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

H0: σ1= σ2= σ3=…… σC = 0 (son iguales) H1: No todas las σj son 0 (no son iguales)

Resultado= Rechazamos la H0, el p-value es menor a 0.05