R Notebook

#librerias
library(tidyr)
library(ggplot2)

set.seed(1000376863)

datos30d = c(10, 12, 13, 14)

bootstrap_30d <- sample(datos30d, size = 500, replace = TRUE)

media_30d <- mean(datos30d)
sd_30d <- sd(datos30d)
normal_30d <- rnorm(500, mean = media_30d, sd = sd_30d)

summary(normal_30d)

   Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
  7.842  11.069  12.372  12.274  13.435  17.212 

summary(bootstrap_30d)

   Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
  10.00   11.50   13.00   12.33   14.00   14.00 

hist(normal_30d, col = "lightblue", main = "Simulación Normal", breaks = 20)

hist(bootstrap_30d, col = "lightgreen", main = "Simulación Bootstrap", breaks = 20)

datos_OG <- data.frame(
A30D = c(25, 27.5, 26.8, 26.2),
A40D = c(32.5, 34.5, 32.5, 31.8),
A50D = c(45.6, 44.5, 43.8, 43.6),
ICOL = c(10, 5, 2.5, 2.5),
SCOL = c(5, 5, 5, 2.5),
IPHY = c(5, 10, 7.5, 5),
SPHY = c(5, 5, 7.5, 5),
ISEP = c(10, 12.5, 12.5, 10),
SSEP = c(10, 10, 10, 12.5),
ICER = c(5, 10, 10, 10),
SCER = c(5, 2.5, 5, 2.5),
testA30D = c(25.6, 25.4, 25.8, 26),
testA40D = c(29.5, 30.5, 30.8, 29.5),
testA50D = c(39.6, 40, 40.6, 40.7),
testICOL = c(22.5, 15.8, 20, 22.5),
testSCOL = c(15, 15, 17.5, 15),
testIPHY = c(25, 20, 25, 25),
testSPHY = c(20, 22.5, 22.5, 25),
testISEP = c(22.5, 22.5, 20, 20),
testSSEP = c(15, 17.5, 20, 5),
testICER = c(15, 15, 20, 20),
testSCER = c(15, 12.5, 15, 20)
)
head(datos_OG)

simulaciones <- data.frame(
A30D = rnorm(500, mean = mean(datos_OG$A30D), sd = sd(datos_OG$A30D)),
A40D = rnorm(500, mean = mean(datos_OG$A40D), sd = sd(datos_OG$A40D)),
A50D = rnorm(500, mean = mean(datos_OG$A50D), sd = sd(datos_OG$A50D)),
ICOL= rnorm(500, mean = mean(datos_OG$ICOL), sd = sd(datos_OG$ICOL)),
SCOL = rnorm(500, mean = mean(datos_OG$SCOL), sd = sd(datos_OG$SCOL)),
IPHY = rnorm(500, mean = mean(datos_OG$IPHY), sd = sd(datos_OG$IPHY)),
SPHY = rnorm(500, mean = mean(datos_OG$SPHY), sd = sd(datos_OG$SPHY)),
ISEP = rnorm(500, mean = mean(datos_OG$ISEP), sd = sd(datos_OG$ISEP)),
SSEP = rnorm(500, mean = mean(datos_OG$SSEP), sd = sd(datos_OG$SSEP)),
ICER = rnorm(500, mean = mean(datos_OG$ICER), sd = sd(datos_OG$ICER)),
SCER = rnorm(500, mean = mean(datos_OG$SCER), sd = sd(datos_OG$SCER)),
testA30D = rnorm(500, mean = mean(datos_OG$testA30D), sd = sd(datos_OG$testA30D)),
testA40D = rnorm(500, mean = mean(datos_OG$testA40D), sd = sd(datos_OG$testA40D)),
testA50D = rnorm(500, mean = mean(datos_OG$testA50D), sd = sd(datos_OG$testA50D)),
testICOL = rnorm(500, mean = mean(datos_OG$testICOL), sd = sd(datos_OG$testICOL)),
testSCOL = rnorm(500, mean = mean(datos_OG$testSCOL), sd = sd(datos_OG$testSCOL)),
testIPHY = rnorm(500, mean = mean(datos_OG$testIPHY), sd = sd(datos_OG$testIPHY)),
testSPHY = rnorm(500, mean = mean(datos_OG$testSPHY), sd = sd(datos_OG$testSPHY)),
testISEP = rnorm(500, mean = mean(datos_OG$testISEP), sd = sd(datos_OG$testISEP)),
testSSEP = rnorm(500, mean = mean(datos_OG$testSSEP), sd = sd(datos_OG$testSSEP)),
testICER = rnorm(500, mean = mean(datos_OG$testICER), sd = sd(datos_OG$testICER)),
testSCER = rnorm(500, mean = mean(datos_OG$testSCER), sd = sd(datos_OG$testSCER))
)
head(simulaciones)

dim(datos_OG)

[1]  4 22

dim(simulaciones)

[1] 500  22

summary(datos_OG$A30D)

   Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
  25.00   25.90   26.50   26.38   26.98   27.50 

summary(simulaciones$A30D)

   Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
  23.21   25.66   26.47   26.42   27.18   29.73 

simulaciones_lg <- pivot_longer(simulaciones, cols = everything(), names_to = "Tratamiento", values_to = "Altura")

ggplot(simulaciones_lg, aes(x = Altura, fill = Tratamiento)) +
  geom_histogram(alpha = 0.5, position = "identity", bins = 30) +
  facet_wrap(~ Tratamiento, scales = "free") +
  theme_minimal() +
  labs(title = "Simulación de Mediciones para cada tratamiento",
       x = "Medición",
       y = "Frecuencia")

# Función para aplicar la prueba de Shapiro-Wilk a cada columna
normalidad <- sapply(simulaciones, function(x) shapiro.test(x)$p.value)

# Convertir a un data frame para mejor visualización
normalidad_df <- data.frame(Tratamiento = names(normalidad),
                            p_value = normalidad)

# Mostrar los resultados
print(normalidad_df)

# Interpretación: Si p-value > 0.05, no se rechaza la hipótesis nula de normalidad
normalidad_df$Normalidad <- ifelse(normalidad_df$p_value > 0.05, "Sí", "No")

print(normalidad_df)

colnames(simulaciones) <- paste0("T", 1:22)

# Comparación de medias entre columnas 1-11 con sus respectivas en 12-22
comparacion_medias <- data.frame(Columna1 = character(), Columna2 = character(), p_value = numeric())

for (i in 1:11) {
  test <- t.test(simulaciones[[i]], simulaciones[[i + 11]])
  
  comparacion_medias <- rbind(comparacion_medias, 
                              data.frame(Columna1 = colnames(simulaciones)[i],
                                         Columna2 = colnames(simulaciones)[i + 11],
                                         p_value = test$p.value))
}

# Evaluar si hay diferencias significativas (p < 0.05)
comparacion_medias$Diferencia_Significativa <- ifelse(comparacion_medias$p_value < 0.05, "Sí", "No")

# Mostrar resultados
print(comparacion_medias)

comparacion_medias$Variable <- c(
  "Altura (cm) 30 días",
  "Altura (cm) 40 días",
  "Altura (cm) 50 días",
  "Incidencia Colletotrichum",
  "Severidad Colletotrichum",
  "Incidencia Phytophthora",
  "Severidad Phytophthora",
  "Incidencia Septoria",
  "Severidad Septoria",
  "Incidencia Cercospora",
  "Severidad Cercospora"
)