Analisis de covarianza Mice

Análisis de covarianza

library(readxl)
Analisis_de_covarianza_Mice <- read_excel("C:/Users/ERICK/Desktop/Semestre 2021-I/Disenio de Experimentos/Ejercicios/Analisis de covarianza Mice.xlsx")
ff=Analisis_de_covarianza_Mice$Fertilizantes
MO=Analisis_de_covarianza_Mice$MO
Prot=Analisis_de_covarianza_Mice$Proteína
Lat=Analisis_de_covarianza_Mice$X
Long=Analisis_de_covarianza_Mice$Y

#Contar el total de NAs en la base de datos
sum(is.na(Analisis_de_covarianza_Mice))

## [1] 5

#Omitir las filas con observaciones NA
Base_cov <- na.omit(Analisis_de_covarianza_Mice)

#Saber el número de NAs por columna
colSums(is.na(Analisis_de_covarianza_Mice))

##             X             Y Fertilizantes            MO      Proteína 
##             0             0             0             0             5

#install.packages("mice", dependencies = TRUE)
library(mice)

## 
## Attaching package: 'mice'

## The following object is masked from 'package:stats':
## 
##     filter

## The following objects are masked from 'package:base':
## 
##     cbind, rbind

columns <- c("Proteína","MO")
imputed_data <- mice(Analisis_de_covarianza_Mice[,names(Analisis_de_covarianza_Mice) %in% columns],m = 1,
  maxit = 1, method = "mean",print=F)
complete.data_Erick <- mice::complete(imputed_data)
Prot2=complete.data_Erick$Proteína

Análisis exploratorio

\[y_i = \beta_0 + \beta_1x_i+\epsilon_i\]

\[P_i = \beta_0 + \beta_1 MO_i+\epsilon_i\] \[E[P_i|MO_i] = \beta_0 + \beta_1 MO_i\] \[\widehat{P_i} = \beta_0 + \beta_1 MO_i\]

Ajustando el modelo de recta con: Minimos Cuadrados Oridnarios

mod1 = lm(Prot~MO)
summary(mod1)

## 
## Call:
## lm(formula = Prot ~ MO)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -1.10399 -0.27435  0.09673  0.23745  0.76812 
## 
## Coefficients:
##             Estimate Std. Error t value Pr(>|t|)    
## (Intercept)   3.6768     0.3822   9.619 1.58e-10 ***
## MO            0.7050     0.1250   5.642 4.27e-06 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.4003 on 29 degrees of freedom
##   (5 observations deleted due to missingness)
## Multiple R-squared:  0.5233, Adjusted R-squared:  0.5069 
## F-statistic: 31.83 on 1 and 29 DF,  p-value: 4.269e-06

coef = round(mod1$coefficients, 2)

\[\widehat{P_i} = 3.68 + 0.7 MO_i\]

shapiro.test(mod1$residuals)

## 
##  Shapiro-Wilk normality test
## 
## data:  mod1$residuals
## W = 0.97525, p-value = 0.6723

hist(mod1$residuals)

plot(mod1$residuals, pch = 16)

# Análisis de covarianza

mod2 = aov(Prot2 ~ MO + ff)
summary(mod2)

##             Df Sum Sq Mean Sq F value   Pr(>F)    
## MO           1  4.712   4.712  29.921 5.58e-06 ***
## ff           3  0.153   0.051   0.323    0.809    
## Residuals   31  4.882   0.157                     
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

boxplot(Prot2 ~ ff)

corr.plot(MO, prot)