DCAB.R

#Asignatura: Dise昼㸱os Experimentales 
#Facultad  : Ingenier攼㹤a Estadistica e Informatica
#Universidad Nacional del Altiplano
library(knitr)

## Warning: package 'knitr' was built under R version 3.6.3

library(readxl)

## Warning: package 'readxl' was built under R version 3.6.3

DCAB <- read_excel("DCAB.xlsx")
DCAB

## # A tibble: 16 x 3
##    ph    FrutaA    FrutaB
##    <chr> <chr>     <chr> 
##  1 3.46  Melocoton Piña  
##  2 3.48  Melocoton Piña  
##  3 3.55  Melocoton Piña  
##  4 3.54  Melocoton Piña  
##  5 3.57  Melocoton Fresa 
##  6 3.53  Melocoton Fresa 
##  7 3.40  Melocoton Fresa 
##  8 3.44  Melocoton Fresa 
##  9 3.51  Papaya    Piña  
## 10 3.60  Papaya    Piña  
## 11 3.41  Papaya    Piña  
## 12 3.52  Papaya    Piña  
## 13 3.49  Papaya    Fresa 
## 14 3.41  Papaya    Fresa 
## 15 3.50  Papaya    Fresa 
## 16 3.58  Papaya    Fresa

head(DCAB)

## # A tibble: 6 x 3
##   ph    FrutaA    FrutaB
##   <chr> <chr>     <chr> 
## 1 3.46  Melocoton Piña  
## 2 3.48  Melocoton Piña  
## 3 3.55  Melocoton Piña  
## 4 3.54  Melocoton Piña  
## 5 3.57  Melocoton Fresa 
## 6 3.53  Melocoton Fresa

View(DCAB)

#Factores
DCAB$FrutaA <- factor(DCAB$FrutaA)
DCAB$FrutaB <- factor(DCAB$FrutaB)
DCAB$ph <- as.numeric(DCAB$ph)
str(DCAB)

## tibble [16 x 3] (S3: tbl_df/tbl/data.frame)
##  $ ph    : num [1:16] 3.46 3.48 3.55 3.54 3.57 3.53 3.4 3.44 3.51 3.6 ...
##  $ FrutaA: Factor w/ 2 levels "Melocoton","Papaya": 1 1 1 1 1 1 1 1 2 2 ...
##  $ FrutaB: Factor w/ 2 levels "Fresa","Piña": 2 2 2 2 1 1 1 1 2 2 ...

head(DCAB)

## # A tibble: 6 x 3
##      ph FrutaA    FrutaB
##   <dbl> <fct>     <fct> 
## 1  3.46 Melocoton Piña  
## 2  3.48 Melocoton Piña  
## 3  3.55 Melocoton Piña  
## 4  3.54 Melocoton Piña  
## 5  3.57 Melocoton Fresa 
## 6  3.53 Melocoton Fresa

#Analisis del ANOVA
mod1 <- aov(ph~ FrutaA + FrutaB, data = DCAB )
mod1

## Call:
##    aov(formula = ph ~ FrutaA + FrutaB, data = DCAB)
## 
## Terms:
##                     FrutaA     FrutaB  Residuals
## Sum of Squares  0.00015625 0.00140625 0.05713125
## Deg. of Freedom          1          1         13
## 
## Residual standard error: 0.06629262
## Estimated effects may be unbalanced

summary(mod1)

##             Df  Sum Sq  Mean Sq F value Pr(>F)
## FrutaA       1 0.00016 0.000156   0.036  0.853
## FrutaB       1 0.00141 0.001406   0.320  0.581
## Residuals   13 0.05713 0.004395

coef(mod1)

##  (Intercept) FrutaAPapaya   FrutaBPiña 
##     3.486875     0.006250     0.018750

mod2 <- aov(ph~ FrutaB + FrutaA, data = DCAB )
mod2

## Call:
##    aov(formula = ph ~ FrutaB + FrutaA, data = DCAB)
## 
## Terms:
##                     FrutaB     FrutaA  Residuals
## Sum of Squares  0.00140625 0.00015625 0.05713125
## Deg. of Freedom          1          1         13
## 
## Residual standard error: 0.06629262
## Estimated effects may be unbalanced

summary(mod2)

##             Df  Sum Sq  Mean Sq F value Pr(>F)
## FrutaB       1 0.00141 0.001406   0.320  0.581
## FrutaA       1 0.00016 0.000156   0.036  0.853
## Residuals   13 0.05713 0.004395

coef(mod2)

##  (Intercept)   FrutaBPiña FrutaAPapaya 
##     3.486875     0.018750     0.006250

#Graficas de Cajas
boxplot(ph~FrutaA, data=DCAB, col=c("blue","orange"), main="Diagrama de Cajas de los Jugos de Frutas")
mtext("By:Amarillas", side = 3, adj = 1, family = "mono")

boxplot(ph~FrutaB, data=DCAB, col=c("yellow","green"), main="Diagrama de Cajas de los Jugos de Frutas")
mtext("By:Amarillas", side = 3, adj = 1, family = "mono")

#Hipotesis de Normalidad
shapiro.test(mod1$residuals)

## 
##  Shapiro-Wilk normality test
## 
## data:  mod1$residuals
## W = 0.94645, p-value = 0.4357

#Grafico
qqnorm(mod1$residuals)
mtext("By:Amarillas", side = 3, adj = 1, family = "mono")

#Hipotesis de Homogeneidad
bartlett.test(DCAB$ph, DCAB$FrutaA)

## 
##  Bartlett test of homogeneity of variances
## 
## data:  DCAB$ph and DCAB$FrutaA
## Bartlett's K-squared = 0.11591, df = 1, p-value = 0.7335

bartlett.test(DCAB$ph, DCAB$FrutaB)

## 
##  Bartlett test of homogeneity of variances
## 
## data:  DCAB$ph and DCAB$FrutaB
## Bartlett's K-squared = 0.16743, df = 1, p-value = 0.6824

#Hipotesis de Independencia
layout(matrix(c(1,2,3,4),1,1))
plot(mod1)

plot(mod2)

library(agricolae)

## Warning: package 'agricolae' was built under R version 3.6.3

contraste  <- SNK.test(mod1,"Fruta A", console=TRUE, main="Contraste de Newman-Keuls para el factor nivel de la Fruta A")

## Name:  Fruta A 
##  FrutaA FrutaB