Sys.setlocale("LC_ALL", "es_ES.UTF-8")
[1] "LC_COLLATE=es_ES.UTF-8;LC_CTYPE=es_ES.UTF-8;LC_MONETARY=es_ES.UTF-8;LC_NUMERIC=C;LC_TIME=es_ES.UTF-8"

El objetivo de esta actividad es la aplicación práctica de los conceptos explicados durante la semana 25 del Máster.

1 Actividad 1

1.1 Analisis de wine_quality_red

Con el dataset “wine_quality_red.csv”, ¿Qué factores afectan al pH del vino tinto?. Realizad una regresión multilineal y comentad los resultados.

set.seed(123)

Carga y visualización de datos

wine_red <- read_csv("winequality-red.csv", locale = locale(encoding = "UTF-8"))
attach(wine_red)

head(wine_red) %>% kable (format = "html") %>%
  kable_styling (bootstrap_options = c("striped", "hover", "condensed"),
                 full_width = FALSE,
                 position = "center",
                 font_size= 12)
fixed acidity volatile acidity citric acid residual sugar chlorides free sulfur dioxide total sulfur dioxide density pH sulphates alcohol quality
7.4 0.70 0.00 1.9 0.076 11 34 0.9978 3.51 0.56 9.4 5
7.8 0.88 0.00 2.6 0.098 25 67 0.9968 3.20 0.68 9.8 5
7.8 0.76 0.04 2.3 0.092 15 54 0.9970 3.26 0.65 9.8 5
11.2 0.28 0.56 1.9 0.075 17 60 0.9980 3.16 0.58 9.8 6
7.4 0.70 0.00 1.9 0.076 11 34 0.9978 3.51 0.56 9.4 5
7.4 0.66 0.00 1.8 0.075 13 40 0.9978 3.51 0.56 9.4 5

Sea necesario o no para el algoritmo que vayamos a utilizar, siempre es bueno escalar las variables numéricas, para que su peso en las predicciones no se vea afectado por las escalas de valores de cada variable.

# Reescalmos las variables
wine_red_z = scale(wine_red, center = T, scale = T) %>% as.data.frame()

#Compronamos como ha quedao el reescalado.
head (wine_red_z) %>% kable (format = "html") %>%
  kable_styling (bootstrap_options = c("striped", "hover", "condensed"),
                 full_width = FALSE,
                 position = "center",
                 font_size= 12)
fixed acidity volatile acidity citric acid residual sugar chlorides free sulfur dioxide total sulfur dioxide density pH sulphates alcohol quality
-0.5281944 0.9615758 -1.391037 -0.4530767 -0.2436305 -0.4660467 -0.3790141 0.5580999 1.2882399 -0.5790254 -0.9599458 -0.7875763
-0.2984541 1.9668271 -1.391037 0.0434026 0.2238052 0.8723653 0.6241680 0.0282519 -0.7197081 0.1289101 -0.5845942 -0.7875763
-0.2984541 1.2966596 -1.185699 -0.1693742 0.0963227 -0.0836433 0.2289750 0.1342215 -0.3310730 -0.0480738 -0.5845942 -0.7875763
1.6543385 -1.3840105 1.483689 -0.4530767 -0.2648775 0.1075584 0.4113718 0.6640695 -0.9787982 -0.4610361 -0.5845942 0.4507074
-0.5281944 0.9615758 -1.391037 -0.4530767 -0.2436305 -0.4660467 -0.3790141 0.5580999 1.2882399 -0.5790254 -0.9599458 -0.7875763
-0.5281944 0.7381867 -1.391037 -0.5240023 -0.2648775 -0.2748450 -0.1966174 0.5580999 1.2882399 -0.5790254 -0.9599458 -0.7875763

A partir de ahora, vamos a empezar a generar modelos predictivos. Estos modelos necesitan primero ser “entrenados” usando unos datos con resultado conocido. Una vez “entrenado”, para verificar que el modelo producido es bueno, necesitamos “validarlo” usando unos datos con resultado conocido diferentes a los usados para el entrenamiento. Por esta razón vamos a dividir todos los resultados de la tabla en dos grupos, el de entrenamiento (80%) y el de validación (20%)

# Dividimos los datos en "train" (entrenamiento) y "test" (validación)
train = createDataPartition(wine_red_z$pH, p = 0.8, list = FALSE)

train.data = wine_red_z[train, ]
test.data = wine_red_z[-train, ]

filas80pct = nrow (train.data)

filas20pct = nrow (test.data)

La cantidad de filas que conforman el 80% del dataset wine_red_z es de 1281, mientras que el otro 20% es de 318

NOTA: La función createDataPartition trata de mantener en la partición la distribución de la variable dependiente.

  • OBJETIVO: El objetivo principal es modelar la relación entre la variable dependiente (pH) y varias variables independientes (otras variables del dataset) para entender cómo cada variable independiente afecta el pH.
  • CREACIÓN DEL MODELO: Se utiliza un modelo lineal múltiple, que es una extensión del modelo lineal simple. Mientras que en el modelo lineal simple se tiene una sola variable independiente, en el múltiple se tienen varias. La forma general de este modelo es:

Ecuación:
\[** y = b_0 + b_1 \times x_1 + b_2 \times x_2 + b_3 \times x_3 + \dots + b_n \times x_n + e **\]

Donde:
- b0 es el intercepto.
- b1, b2, … , bn son los coeficientes de las variables independientes x1, x2, … , xn.
- e es el término de error.

  • AJUSTE DEL MODELO: Se ajusta el modelo utilizando los datos disponibles. Esto implica estimar los valores de los coeficientes b que mejor se ajustan a tus datos.

Estudio de Multicolinealidad: Factor de Inflación de la Varianza (VIF)

  1. ¿Qué es la Multicolinealidad?: La multicolinealidad ocurre cuando las variables independientes en un modelo de regresión lineal están correlacionadas. Esto puede ser problemático porque puede hacer que la estimación de los coeficientes sea menos precisa, lo que a su vez puede afectar a la interpretación del modelo.

  2. Uso del VIF: El VIF mide cuánto se infla la varianza de un coeficiente estimado debido a la multicolinealidad. Un VIF alto para una variable independiente indica una fuerte correlación con otras variables independientes en el modelo, lo que sugiere multicolinealidad.

  1. Interpretación del VIF:

    • Un VIF de 1 indica que no hay multicolinealidad.
    • Un VIF entre 1 y 5 suele considerarse aceptable.
    • Un VIF mayor a 5 o 10 sugiere una multicolinealidad problemática que necesita ser abordada.

  2. Cómo se Calcula: Para cada variable independiente en el modelo, se ajusta un modelo de regresión lineal usando esa variable como dependiente y las demás independientes del modelo original como predictoras. El VIF se calcula como 1/(1-R²), donde R² es el coeficiente de determinación del modelo ajustado.

  3. Acciones a Tomar en Caso de Multicolinealidad Alta:

    • Revisar las variables y considerar eliminar aquellas con VIF alto.
    • Combinar variables correlacionadas en una sola.
    • Utilizar métodos de regularización como la regresión de Ridge o Lasso.

VAMOS A CREAR EL MODELO DE REGRESIÓN LINEAL

# Primer Modelo
lm1 = lm(pH ~ . - quality, train.data)# pH marca la variable dependiente, "." nos indica que usemos todas las variables independientes, "- quality" quitamos esta variable del análisis.

vif (lm1)
       `fixed acidity`     `volatile acidity`          `citric acid` 
              3.796467               1.815624               3.237577 
      `residual sugar`              chlorides  `free sulfur dioxide` 
              1.523592               1.371009               1.902456 
`total sulfur dioxide`                density              sulphates 
              2.118467               4.025800               1.379428 
               alcohol 
              2.167047

Según este estadístico las variables pueden mostrar multicolinealidad con otras variables cuando el valor es > 5. Como se puede observar ninguna de las variables de estudio muestran valores por encima de 5. Se puede pensar que la variable densidad(4.026) puede presentar multicolinealidad al estar próxima a 5, por lo que para ver si el modelo mejora se elimina esta variable y se comparan los valores de AIC.

# Generamos el modelo 2, quitando la variable densidad
lm2 = update (lm1, . ~ . - density)

# Comparamos los AIC de cada modelo para ver cual es el mejor.
AIC (lm1, lm2)
    df      AIC
lm1 12 2099.120
lm2 11 2624.463

Como se puede ver en los resultados anteriores el modelo lm2 (sin la variable densidad) muestra un valor de AIC de 2624.463 que es mayor que el valor de AIC del modelo con todas las variables (lm1), por lo tanto, se puede asumir que el modelo no mejora al eliminar la variable densidad.
El siguiente paso es examinar los residuos del modelo para comprobar si se cumplen los supuestos necesarios de los modelos lineales.

par (mfrow = c(2, 2) ) #En R se pueden hacer varios graficos por ventana. Para ello, antes de ejecutar la función plot(), se puede ejecutar: par(mfrov = c(filas, columnas) ) Los graficos se irdn mostrando en pantalla por filas. En caso de que se quieran mostrar por columnas en la función anterior se sustituye mfrov por mfcol.

plot (lm1, add.smooth = F)

Supuestos que se deben cumplir para un modelo de regresión lineal

1 No hay Patrones en los Residuos
- Significado. Los residuos de un modelos de regresión lineal son la diferencia entre los valores observados y los valores predichos por el modelo. Idealmente, estos residuos deben ser aleatorios y no mostrar ningún patrón específico. - Cómo se Verifica en Gráficos. Un gráfico de residuos (residuos vs. valores ajustados) no debe mostrar patrones claros. Si los residuos se distribuyen aleatoriamente y no forman un patrón discernible, esto sugiere que el modelo se ajusta bien a los datos. - Importancia. Los patrones en los residuos pueden indicar que el modelo no captura alguna relación no lineal o que existen variables omitidas que son importantes para el modelo.

2 Distribución normal de los Residuos - Significado. Para la mayoría de las pruebas estadísticas en regresión lineal, se asume que los residuos siguen una distribución normal. - Cómo se Verifica en Gráficos. Un gráfico de normalidad, como un histograma de residuos o un gráfico Q-Q (quantile-quantile), se utiliza para evaluar esto. En un gráfico Q-Q, si los puntos caen aproximadamente sobre la línea recta diagonal, sugiere una distribución normal. - Importancia. La normalidad de los residuos asegura que las estimaciones de los intervalos de confianza y las pruebas de hipótesis sean válidas.

3 No hay Heterocedasticidad - Significado La heterocedasticidad ocurre cuando la varianza de los residuos no es constante a través de todas las observaciones. Idealmente, queremos homocedasticidad, donde la varianza de los residuos es constante. - Cómo se verifica en los gráficos: Se observa en un gráfico de residuos vs valores ajustados(Grafico \(\sqrt{Standarized residuals}\)). Si la dispersión de los residuos es uniforme a lo largo del rango de valores ajustados, indica homocedasticidad. La heterocedasticidad se manifiesta como una dispersión que cambia (por ejemplo, forma de embudo) a lo largo del eje horizontal. - Importancia: La heterocedasticidad puede llevar a estimaciones de los coeficientes menos precisas y a pruebas de significancia poco confiables

4 No hay Valores Extremos

  • Significado: Lo valores extremos o atípicos son observaciones que se desvían significativamente de la tendencia general de los datos.
  • Cómo se verifica en los gráficos: Se utilizan gráficos de residuos o de influencia para identificar valores extremos. Los valores atípicos pueden ser evidentes como puntos que se encuentran lejos del grupo principal de puntos en estos gráficos.
  • Importancia: Los valores extremos pueden tener un impacto desproporcionado en la linea de regresión, afectando la precisión e interpretación del modelo. En algunos casos, es crucial investigar y entender la razón detrás de estos valores atípicos.
summary(lm1)

Call:
lm(formula = pH ~ . - quality, data = train.data)

Residuals:
     Min       1Q   Median       3Q      Max 
-2.08788 -0.33481 -0.00191  0.33516  2.95525 

Coefficients:
                        Estimate Std. Error t value Pr(>|t|)    
(Intercept)            -0.002301   0.015268  -0.151   0.8802    
`fixed acidity`        -1.115664   0.030150 -37.003  < 2e-16 ***
`volatile acidity`      0.035472   0.020481   1.732   0.0835 .  
`citric acid`          -0.025707   0.027419  -0.938   0.3486    
`residual sugar`       -0.239932   0.019225 -12.480  < 2e-16 ***
chlorides              -0.148355   0.018766  -7.906 5.74e-15 ***
`free sulfur dioxide`   0.110217   0.020634   5.342 1.09e-07 ***
`total sulfur dioxide` -0.149583   0.022024  -6.792 1.70e-11 ***
density                 0.803731   0.031602  25.433  < 2e-16 ***
sulphates              -0.091186   0.017858  -5.106 3.79e-07 ***
alcohol                 0.505517   0.022754  22.217  < 2e-16 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Residual standard error: 0.5463 on 1270 degrees of freedom
Multiple R-squared:  0.7028,    Adjusted R-squared:  0.7004 
F-statistic: 300.3 on 10 and 1270 DF,  p-value: < 2.2e-16

Pero si queremos obtener el mismo resumen pero con un formato más presentable podemos usar las posibilidad disponibles aquí:
https://strengejacke.github.io/sjPlot/articles/tab_model_estimates.html

library (sjPlot)
library (sjmisc)
library (sjlabelled)

# Generamos una tabla formateada con las datos generados por el modelo.
# tab_model(lm1, collapse.ci = TRUE)
tab_model(lm1)
  pH
Predictors Estimates CI p
(Intercept) -0.00 -0.03 – 0.03 0.880
fixed acidity -1.12 -1.17 – -1.06 <0.001
volatile acidity 0.04 -0.00 – 0.08 0.084
citric acid -0.03 -0.08 – 0.03 0.349
residual sugar -0.24 -0.28 – -0.20 <0.001
chlorides -0.15 -0.19 – -0.11 <0.001
free sulfur dioxide 0.11 0.07 – 0.15 <0.001
total sulfur dioxide -0.15 -0.19 – -0.11 <0.001
density 0.80 0.74 – 0.87 <0.001
sulphates -0.09 -0.13 – -0.06 <0.001
alcohol 0.51 0.46 – 0.55 <0.001
Observations 1281
R2 / R2 adjusted 0.703 / 0.700

Una vez generado el modelo, pasamos a generar las Predicciones o validarlo

predictions = lm1 %>% predict (test.data)

data.frame (
  R2 = R2 (predictions, test.data$pH),
  RMSE = RMSE (predictions, test.data$pH),
  MAE = MAE (predictions, test.data$pH),
  AIC = AIC (lm1),
  BIC = BIC (lm1)
)
         R2      RMSE       MAE     AIC      BIC
1 0.6875361 0.5660411 0.4333743 2099.12 2160.985

Volver al inicio

1.2 Analisis de wine_quality_white

Ahora nos dan también el dataset “wine_quality_white”. Comparad gráficamente y mediante tests de hipótesis las características de ambos vinos. Intentad que visualmente sea muy claro, es decir, son comparaciones y no se admitirán gráficos individuales o que el lector tenga que ir buscando resultados sueltos por todo el documento. Recordad que todos los gráficos y tablas deben ser autoexplicativos (colores, nombres en los ejes, leyendas, títulos…)

# Cargamos la nueva tabla de datos
wine_white <- read_csv("winequality-white.csv", locale = locale(encoding = "UTF-8"))

# Comprobamos la pinta que tiene la tabla
head (wine_white) %>% kable (format = "html") %>%
  kable_styling (bootstrap_options = c("striped", "hover", "condensed"),
                 full_width = FALSE,
                 position = "center",
                 font_size= 12)
fixed acidity volatile acidity citric acid residual sugar chlorides free sulfur dioxide total sulfur dioxide density pH sulphates alcohol quality
7.0 0.27 0.36 20.7 0.045 45 170 1.0010 3.00 0.45 8.8 6
6.3 0.30 0.34 1.6 0.049 14 132 0.9940 3.30 0.49 9.5 6
8.1 0.28 0.40 6.9 0.050 30 97 0.9951 3.26 0.44 10.1 6
7.2 0.23 0.32 8.5 0.058 47 186 0.9956 3.19 0.40 9.9 6
7.2 0.23 0.32 8.5 0.058 47 186 0.9956 3.19 0.40 9.9 6
8.1 0.28 0.40 6.9 0.050 30 97 0.9951 3.26 0.44 10.1 6

Distribución de las distintas variables en función del tipo de vino

# library(ggplot2)
# library(reshape2)
# library(WRS2)

wine_red$wine = "red" # Anadimos una columna con "red" en todas las filas

wine_white$wine = "white" # Añadimos una columna con "white" en todas las filas

#Comprobamos el cambio introducido
head(wine_white) %>% kable(format = "html") %>%
  kable_styling(bootstrap_options = c("striped","hover","condensed"),
                full_width = FALSE,
                position = "center",
                font_size= 12)
fixed acidity volatile acidity citric acid residual sugar chlorides free sulfur dioxide total sulfur dioxide density pH sulphates alcohol quality wine
7.0 0.27 0.36 20.7 0.045 45 170 1.0010 3.00 0.45 8.8 6 white
6.3 0.30 0.34 1.6 0.049 14 132 0.9940 3.30 0.49 9.5 6 white
8.1 0.28 0.40 6.9 0.050 30 97 0.9951 3.26 0.44 10.1 6 white
7.2 0.23 0.32 8.5 0.058 47 186 0.9956 3.19 0.40 9.9 6 white
7.2 0.23 0.32 8.5 0.058 47 186 0.9956 3.19 0.40 9.9 6 white
8.1 0.28 0.40 6.9 0.050 30 97 0.9951 3.26 0.44 10.1 6 white 
# Unimos ambas tablas en una sola.
wines = rbind(wine_red, wine_white)

# Reorganizamos La tabla en una de tres columnas donde aparezca el tipo de vino, las otras variables una tras otra, el valor de dichas variables. Es decir "fundimos" las columnas de las distinas variables en una sola.

wine_r = melt(wines, id.vars = "wine", measure.vars = 1:12)

# Comprobamos como queda la tabla
wine_r %>% head() %>% kable() %>% kable_styling()
wine variable value
red fixed acidity 7.4
red fixed acidity 7.8
red fixed acidity 7.8
red fixed acidity 11.2
red fixed acidity 7.4
red fixed acidity 7.4

Ahora tenemos una tabla con la que facilmente podemos representar la comparativa de cada una de las variables

ggplot(wine_r, aes(x = wine, y = value)) +
  geom_boxplot() +
  facet_wrap(~ variable, nrow = 3, ncol = 4, scales = "free")

Se observan muchos outliers, por lo que los tests de hipótesis se recomiendan que sean robustos.

Las comillas invertidas (`) se utilizan en este caso para manejar los nombres de las columnas que pueden contener espacios o caracteres especiales. En R, si tienes un nombre de columna como “fixed acidity”, necesitas usar comillas invertidas para referirte a ella en una fórmula o en otras funciones, de lo contrario R lo interpretara como dos objetos separados, “fixed” y “acidity”.
Por lo tanto, “paste0(`variable` wine)” esta asegurando que cada nombre de columna se maneje correctamente en la fórmula, sin importar si contiene espacios o caracteres especiales.

# Recogemos Los nombres de las columnas numéricas en un vector
nombres_columnas <- colnames(wines)[sapply(wines, is.numeric)] # Por si hubiese alguna columna no numerica se aplica el sappLy.

# Creamos un dataframe para almacenar Los resultados.
resultados <- data.frame(
  Variable = nombres_columnas,
  Estimador = numeric(length(nombres_columnas)),
  CI_95_inf = numeric(length(nombres_columnas)),
  CI_95_sup = numeric(length(nombres_columnas)),
  p_valor = numeric(length(nombres_columnas)),
  p_valor_ajustado = numeric(length(nombres_columnas))
  )

# Bucle para aplicar el test a cada variable numérica
for(i in 1:length(nombres_columnas)){
  variable = nombres_columnas[i]
  print(variable)
  f = as.formula(paste0("`", variable, "` ~ wine")) # ES un forma de añadirle las comillas invertidas a las nombres de las variables
  print(f)# solo para ver como queda f
  test = pb2gen(f, data = wines, est = "median")
  resultados$Estimador[i] <- test$test
  resultados$CI_95_inf[i] <- test$conf.int[1]
  resultados$CI_95_sup[i] <- test$conf.int[2]
  resultados$p_valor[i] <- test$p.value
}
[1] "fixed acidity"
`fixed acidity` ~ wine
[1] "volatile acidity"
`volatile acidity` ~ wine
[1] "citric acid"
`citric acid` ~ wine
[1] "residual sugar"
`residual sugar` ~ wine
[1] "chlorides"
chlorides ~ wine
[1] "free sulfur dioxide"
`free sulfur dioxide` ~ wine
[1] "total sulfur dioxide"
`total sulfur dioxide` ~ wine
[1] "density"
density ~ wine
[1] "pH"
pH ~ wine
[1] "sulphates"
sulphates ~ wine
[1] "alcohol"
alcohol ~ wine
[1] "quality"
quality ~ wine
# Añadimos el ajuste de Los valores p
resultados$p_valor_ajustado <- p.adjust(resultados$p_valor, method = "bonferroni")

# Mostramos La tabla

resultados %>%
  kable(format = "html") %>%
  kable_styling(bootstrap_options = c("striped","hover","condensed"),
                full_width = FALSE,
                position = "center",
                font_size= 12)
Variable Estimador CI_95_inf CI_95_sup p_valor p_valor_ajustado
fixed acidity 1.10000 1.000000 1.20000 0.0000000 0.0000000
volatile acidity 0.26000 0.245000 0.27000 0.0000000 0.0000000
citric acid -0.06000 -0.080000 -0.05000 0.0000000 0.0000000
residual sugar -3.00000 -3.250000 -2.80000 0.0000000 0.0000000
chlorides 0.03600 0.035000 0.03750 0.0000000 0.0000000
free sulfur dioxide -20.00000 -21.000000 -19.00000 0.0000000 0.0000000
total sulfur dioxide -96.00000 -99.000000 -94.00000 0.0000000 0.0000000
density 0.00301 0.002835 0.00314 0.0000000 0.0000000
pH 0.13000 0.120000 0.15000 0.0000000 0.0000000
sulphates 0.15000 0.130000 0.16000 0.0000000 0.0000000
alcohol -0.20000 -0.300000 -0.10000 0.0050083 0.0601002
quality 0.00000 0.000000 0.00000 0.9949917 1.0000000

Volver al inicio

2 Actividad 2

Realizad un análisis exploratorio, una regresión logística del dataset “diabetes.csv” e intentad predecir la diabetes. La variable dependiente es “Outcome”.

data <- read_csv("diabetes.csv", locale = locale(encoding = "UTF-8"))

kable(head(data, 5), format = "html", escape = FALSE) %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed"),
                full_width = FALSE,
                position = "center",
                font_size = 12)
Pregnancies Glucose BloodPressure SkinThickness Insulin BMI DiabetesPedigreeFunction Age Outcome
6 148 72 35 0 33.6 0.627 50 1
1 85 66 29 0 26.6 0.351 31 0
8 183 64 0 0 23.3 0.672 32 1
1 89 66 23 94 28.1 0.167 21 0
0 137 40 35 168 43.1 2.288 33 1 
data <- data %>%
  mutate(Outcome = ifelse(Outcome == 1, "Positive", "Negative")) %>% # Cambiamos 1 y 0 por positivo o negativo
  mutate(Outcome = factor(Outcome, levels = c("Positive", "Negative"), labels = c("Positive", "Negative"))) %>%
  relocate(Outcome, .before = Pregnancies)
kable(head(data, 5), format = "html", escape = FALSE) %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed"),
                full_width = FALSE,
                position = "center",
                font_size = 12)
Outcome Pregnancies Glucose BloodPressure SkinThickness Insulin BMI DiabetesPedigreeFunction Age
Positive 6 148 72 35 0 33.6 0.627 50
Negative 1 85 66 29 0 26.6 0.351 31
Positive 8 183 64 0 0 23.3 0.672 32
Negative 1 89 66 23 94 28.1 0.167 21
Positive 0 137 40 35 168 43.1 2.288 33

PRESENTACIÓN DEL MODELO FINAL

A continuación se llevarán a cabo los siguientes pasos:

  1. Selección del Modelo Basado en AIC: Se calculará el AIC (Criterio de Información de Akaike) para cada combinación posible de variables explicativas en el modelo de regresión logística. Esto permitirá identificar la combinación de variables que proporciona el mejor equilibrio entre la complejidad del modelo y el ajuste de los datos.

    El AIC combina dos aspectos importantes: qué tan bien el modelo se ajusta a los datos y cuán complejo es el modelo (cuántos parámetros tiene).
    Un modelo con un mejor ajuste tendrá un AIC más bajo, pero añadir más parámetros (haciéndolo más complejo) puede aumentar el AIC. Así que el AIC busca el modelo que tiene el mejor ajuste con la menor complejidad posible.

    El modelo con el AIC más bajo entre un conjunto de modelos es generalmente el considerado mejor.

  2. Construcción y Ajuste del Modelo Final: Una vez identificada la mejor combinación de variables (aquella que tiene el AIC más bajo), se construirá la fórmula del modelo final. Se mostrará también la lista de todas las combinaciones posibles de variables con sus respectivos valores de AIC, ordenados de menor a mayor.

  3. Resumen y Presentación del Modelo Final: Destacaremos aquellos términos estadísticamente significativos en el modelo seleccionado en base al AIC.

#Creamos LA función personalizada para caloular el AIC de un modelo dado
calc_aic <- function(vars, data) {
  # Construimos la fórmula del modelo como una cadena de texto
  formula <- as.formula (paste ("Outcome ~", paste(vars, collapse = "+")))
  #Ajustamos el modelo de regresión logistica
  model <- glm(formula, data = data, family = binomial () )
  # Calcula el AIC del modelo
  AIC (model)
}

# Lista de variables predictoras
predictores <- c('Pregnancies', 'Glucose', 'BloodPressure', 'SkinThickness', 'Insulin' , 'BMI', 'DiabetesPedigreeFunction', 'Age')

# Generamos todas las combinaciones posibles de las variables predictoras
combinaciones <- list ()
for (i in 1:length (predictores)) {
  combinaciones <- c(combinaciones, combn(predictores, i, simplify = FALSE))
}
print (combinaciones)
[[1]]
[1] "Pregnancies"

[[2]]
[1] "Glucose"

[[3]]
[1] "BloodPressure"

[[4]]
[1] "SkinThickness"

[[5]]
[1] "Insulin"

[[6]]
[1] "BMI"

[[7]]
[1] "DiabetesPedigreeFunction"

[[8]]
[1] "Age"

[[9]]
[1] "Pregnancies" "Glucose"    

[[10]]
[1] "Pregnancies"   "BloodPressure"

[[11]]
[1] "Pregnancies"   "SkinThickness"

[[12]]
[1] "Pregnancies" "Insulin"    

[[13]]
[1] "Pregnancies" "BMI"        

[[14]]
[1] "Pregnancies"              "DiabetesPedigreeFunction"

[[15]]
[1] "Pregnancies" "Age"        

[[16]]
[1] "Glucose"       "BloodPressure"

[[17]]
[1] "Glucose"       "SkinThickness"

[[18]]
[1] "Glucose" "Insulin"

[[19]]
[1] "Glucose" "BMI"    

[[20]]
[1] "Glucose"                  "DiabetesPedigreeFunction"

[[21]]
[1] "Glucose" "Age"    

[[22]]
[1] "BloodPressure" "SkinThickness"

[[23]]
[1] "BloodPressure" "Insulin"      

[[24]]
[1] "BloodPressure" "BMI"          

[[25]]
[1] "BloodPressure"            "DiabetesPedigreeFunction"

[[26]]
[1] "BloodPressure" "Age"          

[[27]]
[1] "SkinThickness" "Insulin"      

[[28]]
[1] "SkinThickness" "BMI"          

[[29]]
[1] "SkinThickness"            "DiabetesPedigreeFunction"

[[30]]
[1] "SkinThickness" "Age"          

[[31]]
[1] "Insulin" "BMI"    

[[32]]
[1] "Insulin"                  "DiabetesPedigreeFunction"

[[33]]
[1] "Insulin" "Age"    

[[34]]
[1] "BMI"                      "DiabetesPedigreeFunction"

[[35]]
[1] "BMI" "Age"

[[36]]
[1] "DiabetesPedigreeFunction" "Age"                     

[[37]]
[1] "Pregnancies"   "Glucose"       "BloodPressure"

[[38]]
[1] "Pregnancies"   "Glucose"       "SkinThickness"

[[39]]
[1] "Pregnancies" "Glucose"     "Insulin"    

[[40]]
[1] "Pregnancies" "Glucose"     "BMI"        

[[41]]
[1] "Pregnancies"              "Glucose"                 
[3] "DiabetesPedigreeFunction"

[[42]]
[1] "Pregnancies" "Glucose"     "Age"        

[[43]]
[1] "Pregnancies"   "BloodPressure" "SkinThickness"

[[44]]
[1] "Pregnancies"   "BloodPressure" "Insulin"      

[[45]]
[1] "Pregnancies"   "BloodPressure" "BMI"          

[[46]]
[1] "Pregnancies"              "BloodPressure"           
[3] "DiabetesPedigreeFunction"

[[47]]
[1] "Pregnancies"   "BloodPressure" "Age"          

[[48]]
[1] "Pregnancies"   "SkinThickness" "Insulin"      

[[49]]
[1] "Pregnancies"   "SkinThickness" "BMI"          

[[50]]
[1] "Pregnancies"              "SkinThickness"           
[3] "DiabetesPedigreeFunction"

[[51]]
[1] "Pregnancies"   "SkinThickness" "Age"          

[[52]]
[1] "Pregnancies" "Insulin"     "BMI"        

[[53]]
[1] "Pregnancies"              "Insulin"                 
[3] "DiabetesPedigreeFunction"

[[54]]
[1] "Pregnancies" "Insulin"     "Age"        

[[55]]
[1] "Pregnancies"              "BMI"                     
[3] "DiabetesPedigreeFunction"

[[56]]
[1] "Pregnancies" "BMI"         "Age"        

[[57]]
[1] "Pregnancies"              "DiabetesPedigreeFunction"
[3] "Age"                     

[[58]]
[1] "Glucose"       "BloodPressure" "SkinThickness"

[[59]]
[1] "Glucose"       "BloodPressure" "Insulin"      

[[60]]
[1] "Glucose"       "BloodPressure" "BMI"          

[[61]]
[1] "Glucose"                  "BloodPressure"           
[3] "DiabetesPedigreeFunction"

[[62]]
[1] "Glucose"       "BloodPressure" "Age"          

[[63]]
[1] "Glucose"       "SkinThickness" "Insulin"      

[[64]]
[1] "Glucose"       "SkinThickness" "BMI"          

[[65]]
[1] "Glucose"                  "SkinThickness"           
[3] "DiabetesPedigreeFunction"

[[66]]
[1] "Glucose"       "SkinThickness" "Age"          

[[67]]
[1] "Glucose" "Insulin" "BMI"    

[[68]]
[1] "Glucose"                  "Insulin"                 
[3] "DiabetesPedigreeFunction"

[[69]]
[1] "Glucose" "Insulin" "Age"    

[[70]]
[1] "Glucose"                  "BMI"                     
[3] "DiabetesPedigreeFunction"

[[71]]
[1] "Glucose" "BMI"     "Age"    

[[72]]
[1] "Glucose"                  "DiabetesPedigreeFunction"
[3] "Age"                     

[[73]]
[1] "BloodPressure" "SkinThickness" "Insulin"      

[[74]]
[1] "BloodPressure" "SkinThickness" "BMI"          

[[75]]
[1] "BloodPressure"            "SkinThickness"           
[3] "DiabetesPedigreeFunction"

[[76]]
[1] "BloodPressure" "SkinThickness" "Age"          

[[77]]
[1] "BloodPressure" "Insulin"       "BMI"          

[[78]]
[1] "BloodPressure"            "Insulin"                 
[3] "DiabetesPedigreeFunction"

[[79]]
[1] "BloodPressure" "Insulin"       "Age"          

[[80]]
[1] "BloodPressure"            "BMI"                     
[3] "DiabetesPedigreeFunction"

[[81]]
[1] "BloodPressure" "BMI"           "Age"          

[[82]]
[1] "BloodPressure"            "DiabetesPedigreeFunction"
[3] "Age"                     

[[83]]
[1] "SkinThickness" "Insulin"       "BMI"          

[[84]]
[1] "SkinThickness"            "Insulin"                 
[3] "DiabetesPedigreeFunction"

[[85]]
[1] "SkinThickness" "Insulin"       "Age"          

[[86]]
[1] "SkinThickness"            "BMI"                     
[3] "DiabetesPedigreeFunction"

[[87]]
[1] "SkinThickness" "BMI"           "Age"          

[[88]]
[1] "SkinThickness"            "DiabetesPedigreeFunction"
[3] "Age"                     

[[89]]
[1] "Insulin"                  "BMI"                     
[3] "DiabetesPedigreeFunction"

[[90]]
[1] "Insulin" "BMI"     "Age"    

[[91]]
[1] "Insulin"                  "DiabetesPedigreeFunction"
[3] "Age"                     

[[92]]
[1] "BMI"                      "DiabetesPedigreeFunction"
[3] "Age"                     

[[93]]
[1] "Pregnancies"   "Glucose"       "BloodPressure" "SkinThickness"

[[94]]
[1] "Pregnancies"   "Glucose"       "BloodPressure" "Insulin"      

[[95]]
[1] "Pregnancies"   "Glucose"       "BloodPressure" "BMI"          

[[96]]
[1] "Pregnancies"              "Glucose"                 
[3] "BloodPressure"            "DiabetesPedigreeFunction"

[[97]]
[1] "Pregnancies"   "Glucose"       "BloodPressure" "Age"          

[[98]]
[1] "Pregnancies"   "Glucose"       "SkinThickness" "Insulin"      

[[99]]
[1] "Pregnancies"   "Glucose"       "SkinThickness" "BMI"          

[[100]]
[1] "Pregnancies"              "Glucose"                 
[3] "SkinThickness"            "DiabetesPedigreeFunction"

[[101]]
[1] "Pregnancies"   "Glucose"       "SkinThickness" "Age"          

[[102]]
[1] "Pregnancies" "Glucose"     "Insulin"     "BMI"        

[[103]]
[1] "Pregnancies"              "Glucose"                 
[3] "Insulin"                  "DiabetesPedigreeFunction"

[[104]]
[1] "Pregnancies" "Glucose"     "Insulin"     "Age"        

[[105]]
[1] "Pregnancies"              "Glucose"                 
[3] "BMI"                      "DiabetesPedigreeFunction"

[[106]]
[1] "Pregnancies" "Glucose"     "BMI"         "Age"        

[[107]]
[1] "Pregnancies"              "Glucose"                 
[3] "DiabetesPedigreeFunction" "Age"                     

[[108]]
[1] "Pregnancies"   "BloodPressure" "SkinThickness" "Insulin"      

[[109]]
[1] "Pregnancies"   "BloodPressure" "SkinThickness" "BMI"          

[[110]]
[1] "Pregnancies"              "BloodPressure"           
[3] "SkinThickness"            "DiabetesPedigreeFunction"

[[111]]
[1] "Pregnancies"   "BloodPressure" "SkinThickness" "Age"          

[[112]]
[1] "Pregnancies"   "BloodPressure" "Insulin"       "BMI"          

[[113]]
[1] "Pregnancies"              "BloodPressure"           
[3] "Insulin"                  "DiabetesPedigreeFunction"

[[114]]
[1] "Pregnancies"   "BloodPressure" "Insulin"       "Age"          

[[115]]
[1] "Pregnancies"              "BloodPressure"           
[3] "BMI"                      "DiabetesPedigreeFunction"

[[116]]
[1] "Pregnancies"   "BloodPressure" "BMI"           "Age"          

[[117]]
[1] "Pregnancies"              "BloodPressure"           
[3] "DiabetesPedigreeFunction" "Age"                     

[[118]]
[1] "Pregnancies"   "SkinThickness" "Insulin"       "BMI"          

[[119]]
[1] "Pregnancies"              "SkinThickness"           
[3] "Insulin"                  "DiabetesPedigreeFunction"

[[120]]
[1] "Pregnancies"   "SkinThickness" "Insulin"       "Age"          

[[121]]
[1] "Pregnancies"              "SkinThickness"           
[3] "BMI"                      "DiabetesPedigreeFunction"

[[122]]
[1] "Pregnancies"   "SkinThickness" "BMI"           "Age"          

[[123]]
[1] "Pregnancies"              "SkinThickness"           
[3] "DiabetesPedigreeFunction" "Age"                     

[[124]]
[1] "Pregnancies"              "Insulin"                 
[3] "BMI"                      "DiabetesPedigreeFunction"

[[125]]
[1] "Pregnancies" "Insulin"     "BMI"         "Age"        

[[126]]
[1] "Pregnancies"              "Insulin"                 
[3] "DiabetesPedigreeFunction" "Age"                     

[[127]]
[1] "Pregnancies"              "BMI"                     
[3] "DiabetesPedigreeFunction" "Age"                     

[[128]]
[1] "Glucose"       "BloodPressure" "SkinThickness" "Insulin"      

[[129]]
[1] "Glucose"       "BloodPressure" "SkinThickness" "BMI"          

[[130]]
[1] "Glucose"                  "BloodPressure"           
[3] "SkinThickness"            "DiabetesPedigreeFunction"

[[131]]
[1] "Glucose"       "BloodPressure" "SkinThickness" "Age"          

[[132]]
[1] "Glucose"       "BloodPressure" "Insulin"       "BMI"          

[[133]]
[1] "Glucose"                  "BloodPressure"           
[3] "Insulin"                  "DiabetesPedigreeFunction"

[[134]]
[1] "Glucose"       "BloodPressure" "Insulin"       "Age"          

[[135]]
[1] "Glucose"                  "BloodPressure"           
[3] "BMI"                      "DiabetesPedigreeFunction"

[[136]]
[1] "Glucose"       "BloodPressure" "BMI"           "Age"          

[[137]]
[1] "Glucose"                  "BloodPressure"           
[3] "DiabetesPedigreeFunction" "Age"                     

[[138]]
[1] "Glucose"       "SkinThickness" "Insulin"       "BMI"          

[[139]]
[1] "Glucose"                  "SkinThickness"           
[3] "Insulin"                  "DiabetesPedigreeFunction"

[[140]]
[1] "Glucose"       "SkinThickness" "Insulin"       "Age"          

[[141]]
[1] "Glucose"                  "SkinThickness"           
[3] "BMI"                      "DiabetesPedigreeFunction"

[[142]]
[1] "Glucose"       "SkinThickness" "BMI"           "Age"          

[[143]]
[1] "Glucose"                  "SkinThickness"           
[3] "DiabetesPedigreeFunction" "Age"                     

[[144]]
[1] "Glucose"                  "Insulin"                 
[3] "BMI"                      "DiabetesPedigreeFunction"

[[145]]
[1] "Glucose" "Insulin" "BMI"     "Age"    

[[146]]
[1] "Glucose"                  "Insulin"                 
[3] "DiabetesPedigreeFunction" "Age"                     

[[147]]
[1] "Glucose"                  "BMI"                     
[3] "DiabetesPedigreeFunction" "Age"                     

[[148]]
[1] "BloodPressure" "SkinThickness" "Insulin"       "BMI"          

[[149]]
[1] "BloodPressure"            "SkinThickness"           
[3] "Insulin"                  "DiabetesPedigreeFunction"

[[150]]
[1] "BloodPressure" "SkinThickness" "Insulin"       "Age"          

[[151]]
[1] "BloodPressure"            "SkinThickness"           
[3] "BMI"                      "DiabetesPedigreeFunction"

[[152]]
[1] "BloodPressure" "SkinThickness" "BMI"           "Age"          

[[153]]
[1] "BloodPressure"            "SkinThickness"           
[3] "DiabetesPedigreeFunction" "Age"                     

[[154]]
[1] "BloodPressure"            "Insulin"                 
[3] "BMI"                      "DiabetesPedigreeFunction"

[[155]]
[1] "BloodPressure" "Insulin"       "BMI"           "Age"          

[[156]]
[1] "BloodPressure"            "Insulin"                 
[3] "DiabetesPedigreeFunction" "Age"                     

[[157]]
[1] "BloodPressure"            "BMI"                     
[3] "DiabetesPedigreeFunction" "Age"                     

[[158]]
[1] "SkinThickness"            "Insulin"                 
[3] "BMI"                      "DiabetesPedigreeFunction"

[[159]]
[1] "SkinThickness" "Insulin"       "BMI"           "Age"          

[[160]]
[1] "SkinThickness"            "Insulin"                 
[3] "DiabetesPedigreeFunction" "Age"                     

[[161]]
[1] "SkinThickness"            "BMI"                     
[3] "DiabetesPedigreeFunction" "Age"                     

[[162]]
[1] "Insulin"                  "BMI"                     
[3] "DiabetesPedigreeFunction" "Age"                     

[[163]]
[1] "Pregnancies"   "Glucose"       "BloodPressure" "SkinThickness"
[5] "Insulin"      

[[164]]
[1] "Pregnancies"   "Glucose"       "BloodPressure" "SkinThickness"
[5] "BMI"          

[[165]]
[1] "Pregnancies"              "Glucose"                 
[3] "BloodPressure"            "SkinThickness"           
[5] "DiabetesPedigreeFunction"

[[166]]
[1] "Pregnancies"   "Glucose"       "BloodPressure" "SkinThickness"
[5] "Age"          

[[167]]
[1] "Pregnancies"   "Glucose"       "BloodPressure" "Insulin"      
[5] "BMI"          

[[168]]
[1] "Pregnancies"              "Glucose"                 
[3] "BloodPressure"            "Insulin"                 
[5] "DiabetesPedigreeFunction"

[[169]]
[1] "Pregnancies"   "Glucose"       "BloodPressure" "Insulin"      
[5] "Age"          

[[170]]
[1] "Pregnancies"              "Glucose"                 
[3] "BloodPressure"            "BMI"                     
[5] "DiabetesPedigreeFunction"

[[171]]
[1] "Pregnancies"   "Glucose"       "BloodPressure" "BMI"          
[5] "Age"          

[[172]]
[1] "Pregnancies"              "Glucose"                 
[3] "BloodPressure"            "DiabetesPedigreeFunction"
[5] "Age"                     

[[173]]
[1] "Pregnancies"   "Glucose"       "SkinThickness" "Insulin"      
[5] "BMI"          

[[174]]
[1] "Pregnancies"              "Glucose"                 
[3] "SkinThickness"            "Insulin"                 
[5] "DiabetesPedigreeFunction"

[[175]]
[1] "Pregnancies"   "Glucose"       "SkinThickness" "Insulin"      
[5] "Age"          

[[176]]
[1] "Pregnancies"              "Glucose"                 
[3] "SkinThickness"            "BMI"                     
[5] "DiabetesPedigreeFunction"

[[177]]
[1] "Pregnancies"   "Glucose"       "SkinThickness" "BMI"          
[5] "Age"          

[[178]]
[1] "Pregnancies"              "Glucose"                 
[3] "SkinThickness"            "DiabetesPedigreeFunction"
[5] "Age"                     

[[179]]
[1] "Pregnancies"              "Glucose"                 
[3] "Insulin"                  "BMI"                     
[5] "DiabetesPedigreeFunction"

[[180]]
[1] "Pregnancies" "Glucose"     "Insulin"     "BMI"         "Age"        

[[181]]
[1] "Pregnancies"              "Glucose"                 
[3] "Insulin"                  "DiabetesPedigreeFunction"
[5] "Age"                     

[[182]]
[1] "Pregnancies"              "Glucose"                 
[3] "BMI"                      "DiabetesPedigreeFunction"
[5] "Age"                     

[[183]]
[1] "Pregnancies"   "BloodPressure" "SkinThickness" "Insulin"      
[5] "BMI"          

[[184]]
[1] "Pregnancies"              "BloodPressure"           
[3] "SkinThickness"            "Insulin"                 
[5] "DiabetesPedigreeFunction"

[[185]]
[1] "Pregnancies"   "BloodPressure" "SkinThickness" "Insulin"      
[5] "Age"          

[[186]]
[1] "Pregnancies"              "BloodPressure"           
[3] "SkinThickness"            "BMI"                     
[5] "DiabetesPedigreeFunction"

[[187]]
[1] "Pregnancies"   "BloodPressure" "SkinThickness" "BMI"          
[5] "Age"          

[[188]]
[1] "Pregnancies"              "BloodPressure"           
[3] "SkinThickness"            "DiabetesPedigreeFunction"
[5] "Age"                     

[[189]]
[1] "Pregnancies"              "BloodPressure"           
[3] "Insulin"                  "BMI"                     
[5] "DiabetesPedigreeFunction"

[[190]]
[1] "Pregnancies"   "BloodPressure" "Insulin"       "BMI"          
[5] "Age"          

[[191]]
[1] "Pregnancies"              "BloodPressure"           
[3] "Insulin"                  "DiabetesPedigreeFunction"
[5] "Age"                     

[[192]]
[1] "Pregnancies"              "BloodPressure"           
[3] "BMI"                      "DiabetesPedigreeFunction"
[5] "Age"                     

[[193]]
[1] "Pregnancies"              "SkinThickness"           
[3] "Insulin"                  "BMI"                     
[5] "DiabetesPedigreeFunction"

[[194]]
[1] "Pregnancies"   "SkinThickness" "Insulin"       "BMI"          
[5] "Age"          

[[195]]
[1] "Pregnancies"              "SkinThickness"           
[3] "Insulin"                  "DiabetesPedigreeFunction"
[5] "Age"                     

[[196]]
[1] "Pregnancies"              "SkinThickness"           
[3] "BMI"                      "DiabetesPedigreeFunction"
[5] "Age"                     

[[197]]
[1] "Pregnancies"              "Insulin"                 
[3] "BMI"                      "DiabetesPedigreeFunction"
[5] "Age"                     

[[198]]
[1] "Glucose"       "BloodPressure" "SkinThickness" "Insulin"      
[5] "BMI"          

[[199]]
[1] "Glucose"                  "BloodPressure"           
[3] "SkinThickness"            "Insulin"                 
[5] "DiabetesPedigreeFunction"

[[200]]
[1] "Glucose"       "BloodPressure" "SkinThickness" "Insulin"      
[5] "Age"          

[[201]]
[1] "Glucose"                  "BloodPressure"           
[3] "SkinThickness"            "BMI"                     
[5] "DiabetesPedigreeFunction"

[[202]]
[1] "Glucose"       "BloodPressure" "SkinThickness" "BMI"          
[5] "Age"          

[[203]]
[1] "Glucose"                  "BloodPressure"           
[3] "SkinThickness"            "DiabetesPedigreeFunction"
[5] "Age"                     

[[204]]
[1] "Glucose"                  "BloodPressure"           
[3] "Insulin"                  "BMI"                     
[5] "DiabetesPedigreeFunction"

[[205]]
[1] "Glucose"       "BloodPressure" "Insulin"       "BMI"          
[5] "Age"          

[[206]]
[1] "Glucose"                  "BloodPressure"           
[3] "Insulin"                  "DiabetesPedigreeFunction"
[5] "Age"                     

[[207]]
[1] "Glucose"                  "BloodPressure"           
[3] "BMI"                      "DiabetesPedigreeFunction"
[5] "Age"                     

[[208]]
[1] "Glucose"                  "SkinThickness"           
[3] "Insulin"                  "BMI"                     
[5] "DiabetesPedigreeFunction"

[[209]]
[1] "Glucose"       "SkinThickness" "Insulin"       "BMI"          
[5] "Age"          

[[210]]
[1] "Glucose"                  "SkinThickness"           
[3] "Insulin"                  "DiabetesPedigreeFunction"
[5] "Age"                     

[[211]]
[1] "Glucose"                  "SkinThickness"           
[3] "BMI"                      "DiabetesPedigreeFunction"
[5] "Age"                     

[[212]]
[1] "Glucose"                  "Insulin"                 
[3] "BMI"                      "DiabetesPedigreeFunction"
[5] "Age"                     

[[213]]
[1] "BloodPressure"            "SkinThickness"           
[3] "Insulin"                  "BMI"                     
[5] "DiabetesPedigreeFunction"

[[214]]
[1] "BloodPressure" "SkinThickness" "Insulin"       "BMI"          
[5] "Age"          

[[215]]
[1] "BloodPressure"            "SkinThickness"           
[3] "Insulin"                  "DiabetesPedigreeFunction"
[5] "Age"                     

[[216]]
[1] "BloodPressure"            "SkinThickness"           
[3] "BMI"                      "DiabetesPedigreeFunction"
[5] "Age"                     

[[217]]
[1] "BloodPressure"            "Insulin"                 
[3] "BMI"                      "DiabetesPedigreeFunction"
[5] "Age"                     

[[218]]
[1] "SkinThickness"            "Insulin"                 
[3] "BMI"                      "DiabetesPedigreeFunction"
[5] "Age"                     

[[219]]
[1] "Pregnancies"   "Glucose"       "BloodPressure" "SkinThickness"
[5] "Insulin"       "BMI"          

[[220]]
[1] "Pregnancies"              "Glucose"                 
[3] "BloodPressure"            "SkinThickness"           
[5] "Insulin"                  "DiabetesPedigreeFunction"

[[221]]
[1] "Pregnancies"   "Glucose"       "BloodPressure" "SkinThickness"
[5] "Insulin"       "Age"          

[[222]]
[1] "Pregnancies"              "Glucose"                 
[3] "BloodPressure"            "SkinThickness"           
[5] "BMI"                      "DiabetesPedigreeFunction"

[[223]]
[1] "Pregnancies"   "Glucose"       "BloodPressure" "SkinThickness"
[5] "BMI"           "Age"          

[[224]]
[1] "Pregnancies"              "Glucose"                 
[3] "BloodPressure"            "SkinThickness"           
[5] "DiabetesPedigreeFunction" "Age"                     

[[225]]
[1] "Pregnancies"              "Glucose"                 
[3] "BloodPressure"            "Insulin"                 
[5] "BMI"                      "DiabetesPedigreeFunction"

[[226]]
[1] "Pregnancies"   "Glucose"       "BloodPressure" "Insulin"      
[5] "BMI"           "Age"          

[[227]]
[1] "Pregnancies"              "Glucose"                 
[3] "BloodPressure"            "Insulin"                 
[5] "DiabetesPedigreeFunction" "Age"                     

[[228]]
[1] "Pregnancies"              "Glucose"                 
[3] "BloodPressure"            "BMI"                     
[5] "DiabetesPedigreeFunction" "Age"                     

[[229]]
[1] "Pregnancies"              "Glucose"                 
[3] "SkinThickness"            "Insulin"                 
[5] "BMI"                      "DiabetesPedigreeFunction"

[[230]]
[1] "Pregnancies"   "Glucose"       "SkinThickness" "Insulin"      
[5] "BMI"           "Age"          

[[231]]
[1] "Pregnancies"              "Glucose"                 
[3] "SkinThickness"            "Insulin"                 
[5] "DiabetesPedigreeFunction" "Age"                     

[[232]]
[1] "Pregnancies"              "Glucose"                 
[3] "SkinThickness"            "BMI"                     
[5] "DiabetesPedigreeFunction" "Age"                     

[[233]]
[1] "Pregnancies"              "Glucose"                 
[3] "Insulin"                  "BMI"                     
[5] "DiabetesPedigreeFunction" "Age"                     

[[234]]
[1] "Pregnancies"              "BloodPressure"           
[3] "SkinThickness"            "Insulin"                 
[5] "BMI"                      "DiabetesPedigreeFunction"

[[235]]
[1] "Pregnancies"   "BloodPressure" "SkinThickness" "Insulin"      
[5] "BMI"           "Age"          

[[236]]
[1] "Pregnancies"              "BloodPressure"           
[3] "SkinThickness"            "Insulin"                 
[5] "DiabetesPedigreeFunction" "Age"                     

[[237]]
[1] "Pregnancies"              "BloodPressure"           
[3] "SkinThickness"            "BMI"                     
[5] "DiabetesPedigreeFunction" "Age"                     

[[238]]
[1] "Pregnancies"              "BloodPressure"           
[3] "Insulin"                  "BMI"                     
[5] "DiabetesPedigreeFunction" "Age"                     

[[239]]
[1] "Pregnancies"              "SkinThickness"           
[3] "Insulin"                  "BMI"                     
[5] "DiabetesPedigreeFunction" "Age"                     

[[240]]
[1] "Glucose"                  "BloodPressure"           
[3] "SkinThickness"            "Insulin"                 
[5] "BMI"                      "DiabetesPedigreeFunction"

[[241]]
[1] "Glucose"       "BloodPressure" "SkinThickness" "Insulin"      
[5] "BMI"           "Age"          

[[242]]
[1] "Glucose"                  "BloodPressure"           
[3] "SkinThickness"            "Insulin"                 
[5] "DiabetesPedigreeFunction" "Age"                     

[[243]]
[1] "Glucose"                  "BloodPressure"           
[3] "SkinThickness"            "BMI"                     
[5] "DiabetesPedigreeFunction" "Age"                     

[[244]]
[1] "Glucose"                  "BloodPressure"           
[3] "Insulin"                  "BMI"                     
[5] "DiabetesPedigreeFunction" "Age"                     

[[245]]
[1] "Glucose"                  "SkinThickness"           
[3] "Insulin"                  "BMI"                     
[5] "DiabetesPedigreeFunction" "Age"                     

[[246]]
[1] "BloodPressure"            "SkinThickness"           
[3] "Insulin"                  "BMI"                     
[5] "DiabetesPedigreeFunction" "Age"                     

[[247]]
[1] "Pregnancies"              "Glucose"                 
[3] "BloodPressure"            "SkinThickness"           
[5] "Insulin"                  "BMI"                     
[7] "DiabetesPedigreeFunction"

[[248]]
[1] "Pregnancies"   "Glucose"       "BloodPressure" "SkinThickness"
[5] "Insulin"       "BMI"           "Age"          

[[249]]
[1] "Pregnancies"              "Glucose"                 
[3] "BloodPressure"            "SkinThickness"           
[5] "Insulin"                  "DiabetesPedigreeFunction"
[7] "Age"                     

[[250]]
[1] "Pregnancies"              "Glucose"                 
[3] "BloodPressure"            "SkinThickness"           
[5] "BMI"                      "DiabetesPedigreeFunction"
[7] "Age"                     

[[251]]
[1] "Pregnancies"              "Glucose"                 
[3] "BloodPressure"            "Insulin"                 
[5] "BMI"                      "DiabetesPedigreeFunction"
[7] "Age"                     

[[252]]
[1] "Pregnancies"              "Glucose"                 
[3] "SkinThickness"            "Insulin"                 
[5] "BMI"                      "DiabetesPedigreeFunction"
[7] "Age"                     

[[253]]
[1] "Pregnancies"              "BloodPressure"           
[3] "SkinThickness"            "Insulin"                 
[5] "BMI"                      "DiabetesPedigreeFunction"
[7] "Age"                     

[[254]]
[1] "Glucose"                  "BloodPressure"           
[3] "SkinThickness"            "Insulin"                 
[5] "BMI"                      "DiabetesPedigreeFunction"
[7] "Age"                     

[[255]]
[1] "Pregnancies"              "Glucose"                 
[3] "BloodPressure"            "SkinThickness"           
[5] "Insulin"                  "BMI"                     
[7] "DiabetesPedigreeFunction" "Age"

Calculamos ahora el AIC para cada comlanación de variables

# Calculamos ahora el AIC para cada comlanación de variables
aic_vals <- sapply (combinaciones, calc_aic, data = data)
aic_vals
  [1] 960.2099 812.7196 994.1276 993.1890 984.8104 924.7142 974.8609 954.7203
  [9] 790.9500 961.2553 955.3627 945.3401 887.1438 936.4855 947.0981 814.5946
 [17] 813.0668 813.7694 777.4030 802.9874 803.3621 993.0831 984.4124 926.6531
 [25] 974.1017 956.6631 986.4636 924.8473 975.3182 948.1391 922.5638 969.0193
 [33] 941.0204 912.5037 881.6754 935.1037 792.0621 789.9204 792.7769 752.1249
 [41] 779.2744 792.0834 957.1700 946.9929 887.3348 937.9182 949.0640 946.4545
 [49] 888.5193 935.4908 940.2193 882.5455 927.4275 931.7618 872.8374 873.3333
 [57] 925.2260 814.6791 815.6765 777.0661 804.8254 804.2896 811.8371 778.2007
 [65] 804.4607 801.9037 775.7865 802.8060 805.1738 770.8709 763.6846 793.6857
 [73] 986.3133 926.8398 975.2200 949.9084 924.4602 968.9094 942.9899 914.4484
 [81] 879.8930 937.0948 918.7780 971.0133 941.2639 910.9413 883.4654 932.7015
 [89] 912.3720 877.7959 926.8503 870.0069 790.0628 793.9369 748.4285 780.2095
 [97] 792.8617 790.1088 753.5999 779.9315 790.5786 752.4279 780.4989 793.9880
[105] 744.3059 752.0995 780.6077 948.2765 888.9702 937.3233 941.9259 882.3473
[113] 929.2218 933.7002 872.9567 870.7854 927.2254 880.5702 929.2752 932.0934
[121] 873.0775 875.1755 922.7824 870.8009 868.4742 915.6898 860.2278 813.3188
[129] 778.2337 806.1477 801.5442 775.4810 804.6953 806.1720 770.5636 759.5887
[137] 794.5402 777.7229 802.2643 801.8169 770.5938 765.3455 794.0058 767.5500
[145] 763.9777 794.8213 757.2337 920.7755 970.8758 943.0419 912.9413 881.8709
[153] 934.5315 914.2895 875.4573 928.8018 868.3643 907.1214 877.3072 928.1656
[161] 871.0400 868.4942 790.0907 750.2982 780.1796 789.9463 748.8900 781.5465
[169] 794.8170 740.5596 746.9861 781.2408 754.3985 779.4930 790.9478 745.0586
[177] 753.7527 780.9795 743.5062 752.7533 781.9854 744.5088 881.0546 931.1239
[185] 933.8083 873.6613 872.7834 924.5488 870.6060 865.2376 917.5926 857.7617
[193] 867.4738 867.9545 917.1144 861.3496 857.8949 777.4792 803.8132 801.3315
[201] 770.8049 761.5588 793.9574 767.2653 760.1704 795.8361 753.2664 769.2725
[209] 765.9777 793.1387 758.2431 756.2950 909.1206 875.8138 930.0006 869.8990
[217] 866.4310 866.9614 750.8479 779.5629 790.1895 742.0039 748.9714 780.6210
[225] 739.9706 747.9176 782.7657 739.4617 745.3142 754.7446 780.7205 745.5228
[233] 744.1287 868.0758 865.6726 918.8826 859.4162 854.8692 856.2389 769.1649
[241] 762.0105 792.9581 754.9180 752.6929 758.2216 865.8970 741.9676 749.7849
[249] 780.2247 741.1871 739.4534 745.9930 854.3724 754.6780 741.4454

Seleccionamos la combinación con el AIC más bajo

# Seleccionamos la combinación con el AIC más bajo
mejor_modelo <- combinaciones [[which.min(aic_vals) ]]
mejor_modelo
[1] "Pregnancies"              "Glucose"                 
[3] "BloodPressure"            "Insulin"                 
[5] "BMI"                      "DiabetesPedigreeFunction"
[7] "Age"

Construimos la formula de mejor modelo

# Construimos la formula de mejor modelo
formula_mejor_modelo <- paste ("Outcome ~", paste (mejor_modelo, collapse = "+") )

Ajusta el modelo final con la mejor combinación de variables

# Ajusta el modelo final con la mejor combinación de variables
modelo_final <- glm(as.formula(formula_mejor_modelo), data = data, family = binomial())

Crea un dataframe que incluye la formula de todos los modelos y el AIC correspondiente

# Crea un dataframe que incluye la formula de todos los modelos y el AIC correspondiente
modelos_aic <- data.frame (
  Formula = sapply (combinaciones, function(vars) paste("Outcome ~", paste(vars, collapse = "+"))),
  AIC = aic_vals
)

kable(modelos_aic, format = "html", escape = FALSE) %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed"),
                full_width = FALSE,
                position = "center",
                font_size = 12)
Formula AIC
Outcome ~ Pregnancies 960.2099
Outcome ~ Glucose 812.7196
Outcome ~ BloodPressure 994.1276
Outcome ~ SkinThickness 993.1890
Outcome ~ Insulin 984.8104
Outcome ~ BMI 924.7142
Outcome ~ DiabetesPedigreeFunction 974.8609
Outcome ~ Age 954.7203
Outcome ~ Pregnancies+Glucose 790.9500
Outcome ~ Pregnancies+BloodPressure 961.2553
Outcome ~ Pregnancies+SkinThickness 955.3627
Outcome ~ Pregnancies+Insulin 945.3401
Outcome ~ Pregnancies+BMI 887.1438
Outcome ~ Pregnancies+DiabetesPedigreeFunction 936.4855
Outcome ~ Pregnancies+Age 947.0981
Outcome ~ Glucose+BloodPressure 814.5946
Outcome ~ Glucose+SkinThickness 813.0668
Outcome ~ Glucose+Insulin 813.7694
Outcome ~ Glucose+BMI 777.4030
Outcome ~ Glucose+DiabetesPedigreeFunction 802.9874
Outcome ~ Glucose+Age 803.3621
Outcome ~ BloodPressure+SkinThickness 993.0831
Outcome ~ BloodPressure+Insulin 984.4124
Outcome ~ BloodPressure+BMI 926.6531
Outcome ~ BloodPressure+DiabetesPedigreeFunction 974.1017
Outcome ~ BloodPressure+Age 956.6631
Outcome ~ SkinThickness+Insulin 986.4636
Outcome ~ SkinThickness+BMI 924.8473
Outcome ~ SkinThickness+DiabetesPedigreeFunction 975.3182
Outcome ~ SkinThickness+Age 948.1391
Outcome ~ Insulin+BMI 922.5638
Outcome ~ Insulin+DiabetesPedigreeFunction 969.0193
Outcome ~ Insulin+Age 941.0204
Outcome ~ BMI+DiabetesPedigreeFunction 912.5037
Outcome ~ BMI+Age 881.6754
Outcome ~ DiabetesPedigreeFunction+Age 935.1037
Outcome ~ Pregnancies+Glucose+BloodPressure 792.0621
Outcome ~ Pregnancies+Glucose+SkinThickness 789.9204
Outcome ~ Pregnancies+Glucose+Insulin 792.7769
Outcome ~ Pregnancies+Glucose+BMI 752.1249
Outcome ~ Pregnancies+Glucose+DiabetesPedigreeFunction 779.2744
Outcome ~ Pregnancies+Glucose+Age 792.0834
Outcome ~ Pregnancies+BloodPressure+SkinThickness 957.1700
Outcome ~ Pregnancies+BloodPressure+Insulin 946.9929
Outcome ~ Pregnancies+BloodPressure+BMI 887.3348
Outcome ~ Pregnancies+BloodPressure+DiabetesPedigreeFunction 937.9182
Outcome ~ Pregnancies+BloodPressure+Age 949.0640
Outcome ~ Pregnancies+SkinThickness+Insulin 946.4545
Outcome ~ Pregnancies+SkinThickness+BMI 888.5193
Outcome ~ Pregnancies+SkinThickness+DiabetesPedigreeFunction 935.4908
Outcome ~ Pregnancies+SkinThickness+Age 940.2193
Outcome ~ Pregnancies+Insulin+BMI 882.5455
Outcome ~ Pregnancies+Insulin+DiabetesPedigreeFunction 927.4275
Outcome ~ Pregnancies+Insulin+Age 931.7618
Outcome ~ Pregnancies+BMI+DiabetesPedigreeFunction 872.8374
Outcome ~ Pregnancies+BMI+Age 873.3333
Outcome ~ Pregnancies+DiabetesPedigreeFunction+Age 925.2260
Outcome ~ Glucose+BloodPressure+SkinThickness 814.6791
Outcome ~ Glucose+BloodPressure+Insulin 815.6765
Outcome ~ Glucose+BloodPressure+BMI 777.0661
Outcome ~ Glucose+BloodPressure+DiabetesPedigreeFunction 804.8254
Outcome ~ Glucose+BloodPressure+Age 804.2896
Outcome ~ Glucose+SkinThickness+Insulin 811.8371
Outcome ~ Glucose+SkinThickness+BMI 778.2007
Outcome ~ Glucose+SkinThickness+DiabetesPedigreeFunction 804.4607
Outcome ~ Glucose+SkinThickness+Age 801.9037
Outcome ~ Glucose+Insulin+BMI 775.7865
Outcome ~ Glucose+Insulin+DiabetesPedigreeFunction 802.8060
Outcome ~ Glucose+Insulin+Age 805.1738
Outcome ~ Glucose+BMI+DiabetesPedigreeFunction 770.8709
Outcome ~ Glucose+BMI+Age 763.6846
Outcome ~ Glucose+DiabetesPedigreeFunction+Age 793.6857
Outcome ~ BloodPressure+SkinThickness+Insulin 986.3133
Outcome ~ BloodPressure+SkinThickness+BMI 926.8398
Outcome ~ BloodPressure+SkinThickness+DiabetesPedigreeFunction 975.2200
Outcome ~ BloodPressure+SkinThickness+Age 949.9084
Outcome ~ BloodPressure+Insulin+BMI 924.4602
Outcome ~ BloodPressure+Insulin+DiabetesPedigreeFunction 968.9094
Outcome ~ BloodPressure+Insulin+Age 942.9899
Outcome ~ BloodPressure+BMI+DiabetesPedigreeFunction 914.4484
Outcome ~ BloodPressure+BMI+Age 879.8930
Outcome ~ BloodPressure+DiabetesPedigreeFunction+Age 937.0948
Outcome ~ SkinThickness+Insulin+BMI 918.7780
Outcome ~ SkinThickness+Insulin+DiabetesPedigreeFunction 971.0133
Outcome ~ SkinThickness+Insulin+Age 941.2639
Outcome ~ SkinThickness+BMI+DiabetesPedigreeFunction 910.9413
Outcome ~ SkinThickness+BMI+Age 883.4654
Outcome ~ SkinThickness+DiabetesPedigreeFunction+Age 932.7015
Outcome ~ Insulin+BMI+DiabetesPedigreeFunction 912.3720
Outcome ~ Insulin+BMI+Age 877.7959
Outcome ~ Insulin+DiabetesPedigreeFunction+Age 926.8503
Outcome ~ BMI+DiabetesPedigreeFunction+Age 870.0069
Outcome ~ Pregnancies+Glucose+BloodPressure+SkinThickness 790.0628
Outcome ~ Pregnancies+Glucose+BloodPressure+Insulin 793.9369
Outcome ~ Pregnancies+Glucose+BloodPressure+BMI 748.4285
Outcome ~ Pregnancies+Glucose+BloodPressure+DiabetesPedigreeFunction 780.2095
Outcome ~ Pregnancies+Glucose+BloodPressure+Age 792.8617
Outcome ~ Pregnancies+Glucose+SkinThickness+Insulin 790.1088
Outcome ~ Pregnancies+Glucose+SkinThickness+BMI 753.5999
Outcome ~ Pregnancies+Glucose+SkinThickness+DiabetesPedigreeFunction 779.9315
Outcome ~ Pregnancies+Glucose+SkinThickness+Age 790.5786
Outcome ~ Pregnancies+Glucose+Insulin+BMI 752.4279
Outcome ~ Pregnancies+Glucose+Insulin+DiabetesPedigreeFunction 780.4989
Outcome ~ Pregnancies+Glucose+Insulin+Age 793.9880
Outcome ~ Pregnancies+Glucose+BMI+DiabetesPedigreeFunction 744.3059
Outcome ~ Pregnancies+Glucose+BMI+Age 752.0995
Outcome ~ Pregnancies+Glucose+DiabetesPedigreeFunction+Age 780.6077
Outcome ~ Pregnancies+BloodPressure+SkinThickness+Insulin 948.2765
Outcome ~ Pregnancies+BloodPressure+SkinThickness+BMI 888.9702
Outcome ~ Pregnancies+BloodPressure+SkinThickness+DiabetesPedigreeFunction 937.3233
Outcome ~ Pregnancies+BloodPressure+SkinThickness+Age 941.9259
Outcome ~ Pregnancies+BloodPressure+Insulin+BMI 882.3473
Outcome ~ Pregnancies+BloodPressure+Insulin+DiabetesPedigreeFunction 929.2218
Outcome ~ Pregnancies+BloodPressure+Insulin+Age 933.7002
Outcome ~ Pregnancies+BloodPressure+BMI+DiabetesPedigreeFunction 872.9567
Outcome ~ Pregnancies+BloodPressure+BMI+Age 870.7854
Outcome ~ Pregnancies+BloodPressure+DiabetesPedigreeFunction+Age 927.2254
Outcome ~ Pregnancies+SkinThickness+Insulin+BMI 880.5702
Outcome ~ Pregnancies+SkinThickness+Insulin+DiabetesPedigreeFunction 929.2752
Outcome ~ Pregnancies+SkinThickness+Insulin+Age 932.0934
Outcome ~ Pregnancies+SkinThickness+BMI+DiabetesPedigreeFunction 873.0775
Outcome ~ Pregnancies+SkinThickness+BMI+Age 875.1755
Outcome ~ Pregnancies+SkinThickness+DiabetesPedigreeFunction+Age 922.7824
Outcome ~ Pregnancies+Insulin+BMI+DiabetesPedigreeFunction 870.8009
Outcome ~ Pregnancies+Insulin+BMI+Age 868.4742
Outcome ~ Pregnancies+Insulin+DiabetesPedigreeFunction+Age 915.6898
Outcome ~ Pregnancies+BMI+DiabetesPedigreeFunction+Age 860.2278
Outcome ~ Glucose+BloodPressure+SkinThickness+Insulin 813.3188
Outcome ~ Glucose+BloodPressure+SkinThickness+BMI 778.2337
Outcome ~ Glucose+BloodPressure+SkinThickness+DiabetesPedigreeFunction 806.1477
Outcome ~ Glucose+BloodPressure+SkinThickness+Age 801.5442
Outcome ~ Glucose+BloodPressure+Insulin+BMI 775.4810
Outcome ~ Glucose+BloodPressure+Insulin+DiabetesPedigreeFunction 804.6953
Outcome ~ Glucose+BloodPressure+Insulin+Age 806.1720
Outcome ~ Glucose+BloodPressure+BMI+DiabetesPedigreeFunction 770.5636
Outcome ~ Glucose+BloodPressure+BMI+Age 759.5887
Outcome ~ Glucose+BloodPressure+DiabetesPedigreeFunction+Age 794.5402
Outcome ~ Glucose+SkinThickness+Insulin+BMI 777.7229
Outcome ~ Glucose+SkinThickness+Insulin+DiabetesPedigreeFunction 802.2643
Outcome ~ Glucose+SkinThickness+Insulin+Age 801.8169
Outcome ~ Glucose+SkinThickness+BMI+DiabetesPedigreeFunction 770.5938
Outcome ~ Glucose+SkinThickness+BMI+Age 765.3455
Outcome ~ Glucose+SkinThickness+DiabetesPedigreeFunction+Age 794.0058
Outcome ~ Glucose+Insulin+BMI+DiabetesPedigreeFunction 767.5500
Outcome ~ Glucose+Insulin+BMI+Age 763.9777
Outcome ~ Glucose+Insulin+DiabetesPedigreeFunction+Age 794.8213
Outcome ~ Glucose+BMI+DiabetesPedigreeFunction+Age 757.2337
Outcome ~ BloodPressure+SkinThickness+Insulin+BMI 920.7755
Outcome ~ BloodPressure+SkinThickness+Insulin+DiabetesPedigreeFunction 970.8758
Outcome ~ BloodPressure+SkinThickness+Insulin+Age 943.0419
Outcome ~ BloodPressure+SkinThickness+BMI+DiabetesPedigreeFunction 912.9413
Outcome ~ BloodPressure+SkinThickness+BMI+Age 881.8709
Outcome ~ BloodPressure+SkinThickness+DiabetesPedigreeFunction+Age 934.5315
Outcome ~ BloodPressure+Insulin+BMI+DiabetesPedigreeFunction 914.2895
Outcome ~ BloodPressure+Insulin+BMI+Age 875.4573
Outcome ~ BloodPressure+Insulin+DiabetesPedigreeFunction+Age 928.8018
Outcome ~ BloodPressure+BMI+DiabetesPedigreeFunction+Age 868.3643
Outcome ~ SkinThickness+Insulin+BMI+DiabetesPedigreeFunction 907.1214
Outcome ~ SkinThickness+Insulin+BMI+Age 877.3072
Outcome ~ SkinThickness+Insulin+DiabetesPedigreeFunction+Age 928.1656
Outcome ~ SkinThickness+BMI+DiabetesPedigreeFunction+Age 871.0400
Outcome ~ Insulin+BMI+DiabetesPedigreeFunction+Age 868.4942
Outcome ~ Pregnancies+Glucose+BloodPressure+SkinThickness+Insulin 790.0907
Outcome ~ Pregnancies+Glucose+BloodPressure+SkinThickness+BMI 750.2982
Outcome ~ Pregnancies+Glucose+BloodPressure+SkinThickness+DiabetesPedigreeFunction 780.1796
Outcome ~ Pregnancies+Glucose+BloodPressure+SkinThickness+Age 789.9463
Outcome ~ Pregnancies+Glucose+BloodPressure+Insulin+BMI 748.8900
Outcome ~ Pregnancies+Glucose+BloodPressure+Insulin+DiabetesPedigreeFunction 781.5465
Outcome ~ Pregnancies+Glucose+BloodPressure+Insulin+Age 794.8170
Outcome ~ Pregnancies+Glucose+BloodPressure+BMI+DiabetesPedigreeFunction 740.5596
Outcome ~ Pregnancies+Glucose+BloodPressure+BMI+Age 746.9861
Outcome ~ Pregnancies+Glucose+BloodPressure+DiabetesPedigreeFunction+Age 781.2408
Outcome ~ Pregnancies+Glucose+SkinThickness+Insulin+BMI 754.3985
Outcome ~ Pregnancies+Glucose+SkinThickness+Insulin+DiabetesPedigreeFunction 779.4930
Outcome ~ Pregnancies+Glucose+SkinThickness+Insulin+Age 790.9478
Outcome ~ Pregnancies+Glucose+SkinThickness+BMI+DiabetesPedigreeFunction 745.0586
Outcome ~ Pregnancies+Glucose+SkinThickness+BMI+Age 753.7527
Outcome ~ Pregnancies+Glucose+SkinThickness+DiabetesPedigreeFunction+Age 780.9795
Outcome ~ Pregnancies+Glucose+Insulin+BMI+DiabetesPedigreeFunction 743.5062
Outcome ~ Pregnancies+Glucose+Insulin+BMI+Age 752.7533
Outcome ~ Pregnancies+Glucose+Insulin+DiabetesPedigreeFunction+Age 781.9854
Outcome ~ Pregnancies+Glucose+BMI+DiabetesPedigreeFunction+Age 744.5088
Outcome ~ Pregnancies+BloodPressure+SkinThickness+Insulin+BMI 881.0546
Outcome ~ Pregnancies+BloodPressure+SkinThickness+Insulin+DiabetesPedigreeFunction 931.1239
Outcome ~ Pregnancies+BloodPressure+SkinThickness+Insulin+Age 933.8083
Outcome ~ Pregnancies+BloodPressure+SkinThickness+BMI+DiabetesPedigreeFunction 873.6613
Outcome ~ Pregnancies+BloodPressure+SkinThickness+BMI+Age 872.7834
Outcome ~ Pregnancies+BloodPressure+SkinThickness+DiabetesPedigreeFunction+Age 924.5488
Outcome ~ Pregnancies+BloodPressure+Insulin+BMI+DiabetesPedigreeFunction 870.6060
Outcome ~ Pregnancies+BloodPressure+Insulin+BMI+Age 865.2376
Outcome ~ Pregnancies+BloodPressure+Insulin+DiabetesPedigreeFunction+Age 917.5926
Outcome ~ Pregnancies+BloodPressure+BMI+DiabetesPedigreeFunction+Age 857.7617
Outcome ~ Pregnancies+SkinThickness+Insulin+BMI+DiabetesPedigreeFunction 867.4738
Outcome ~ Pregnancies+SkinThickness+Insulin+BMI+Age 867.9545
Outcome ~ Pregnancies+SkinThickness+Insulin+DiabetesPedigreeFunction+Age 917.1144
Outcome ~ Pregnancies+SkinThickness+BMI+DiabetesPedigreeFunction+Age 861.3496
Outcome ~ Pregnancies+Insulin+BMI+DiabetesPedigreeFunction+Age 857.8949
Outcome ~ Glucose+BloodPressure+SkinThickness+Insulin+BMI 777.4792
Outcome ~ Glucose+BloodPressure+SkinThickness+Insulin+DiabetesPedigreeFunction 803.8132
Outcome ~ Glucose+BloodPressure+SkinThickness+Insulin+Age 801.3315
Outcome ~ Glucose+BloodPressure+SkinThickness+BMI+DiabetesPedigreeFunction 770.8049
Outcome ~ Glucose+BloodPressure+SkinThickness+BMI+Age 761.5588
Outcome ~ Glucose+BloodPressure+SkinThickness+DiabetesPedigreeFunction+Age 793.9574
Outcome ~ Glucose+BloodPressure+Insulin+BMI+DiabetesPedigreeFunction 767.2653
Outcome ~ Glucose+BloodPressure+Insulin+BMI+Age 760.1704
Outcome ~ Glucose+BloodPressure+Insulin+DiabetesPedigreeFunction+Age 795.8361
Outcome ~ Glucose+BloodPressure+BMI+DiabetesPedigreeFunction+Age 753.2664
Outcome ~ Glucose+SkinThickness+Insulin+BMI+DiabetesPedigreeFunction 769.2725
Outcome ~ Glucose+SkinThickness+Insulin+BMI+Age 765.9777
Outcome ~ Glucose+SkinThickness+Insulin+DiabetesPedigreeFunction+Age 793.1387
Outcome ~ Glucose+SkinThickness+BMI+DiabetesPedigreeFunction+Age 758.2431
Outcome ~ Glucose+Insulin+BMI+DiabetesPedigreeFunction+Age 756.2950
Outcome ~ BloodPressure+SkinThickness+Insulin+BMI+DiabetesPedigreeFunction 909.1206
Outcome ~ BloodPressure+SkinThickness+Insulin+BMI+Age 875.8138
Outcome ~ BloodPressure+SkinThickness+Insulin+DiabetesPedigreeFunction+Age 930.0006
Outcome ~ BloodPressure+SkinThickness+BMI+DiabetesPedigreeFunction+Age 869.8990
Outcome ~ BloodPressure+Insulin+BMI+DiabetesPedigreeFunction+Age 866.4310
Outcome ~ SkinThickness+Insulin+BMI+DiabetesPedigreeFunction+Age 866.9614
Outcome ~ Pregnancies+Glucose+BloodPressure+SkinThickness+Insulin+BMI 750.8479
Outcome ~ Pregnancies+Glucose+BloodPressure+SkinThickness+Insulin+DiabetesPedigreeFunction 779.5629
Outcome ~ Pregnancies+Glucose+BloodPressure+SkinThickness+Insulin+Age 790.1895
Outcome ~ Pregnancies+Glucose+BloodPressure+SkinThickness+BMI+DiabetesPedigreeFunction 742.0039
Outcome ~ Pregnancies+Glucose+BloodPressure+SkinThickness+BMI+Age 748.9714
Outcome ~ Pregnancies+Glucose+BloodPressure+SkinThickness+DiabetesPedigreeFunction+Age 780.6210
Outcome ~ Pregnancies+Glucose+BloodPressure+Insulin+BMI+DiabetesPedigreeFunction 739.9706
Outcome ~ Pregnancies+Glucose+BloodPressure+Insulin+BMI+Age 747.9176
Outcome ~ Pregnancies+Glucose+BloodPressure+Insulin+DiabetesPedigreeFunction+Age 782.7657
Outcome ~ Pregnancies+Glucose+BloodPressure+BMI+DiabetesPedigreeFunction+Age 739.4617
Outcome ~ Pregnancies+Glucose+SkinThickness+Insulin+BMI+DiabetesPedigreeFunction 745.3142
Outcome ~ Pregnancies+Glucose+SkinThickness+Insulin+BMI+Age 754.7446
Outcome ~ Pregnancies+Glucose+SkinThickness+Insulin+DiabetesPedigreeFunction+Age 780.7205
Outcome ~ Pregnancies+Glucose+SkinThickness+BMI+DiabetesPedigreeFunction+Age 745.5228
Outcome ~ Pregnancies+Glucose+Insulin+BMI+DiabetesPedigreeFunction+Age 744.1287
Outcome ~ Pregnancies+BloodPressure+SkinThickness+Insulin+BMI+DiabetesPedigreeFunction 868.0758
Outcome ~ Pregnancies+BloodPressure+SkinThickness+Insulin+BMI+Age 865.6726
Outcome ~ Pregnancies+BloodPressure+SkinThickness+Insulin+DiabetesPedigreeFunction+Age 918.8826
Outcome ~ Pregnancies+BloodPressure+SkinThickness+BMI+DiabetesPedigreeFunction+Age 859.4162
Outcome ~ Pregnancies+BloodPressure+Insulin+BMI+DiabetesPedigreeFunction+Age 854.8692
Outcome ~ Pregnancies+SkinThickness+Insulin+BMI+DiabetesPedigreeFunction+Age 856.2389
Outcome ~ Glucose+BloodPressure+SkinThickness+Insulin+BMI+DiabetesPedigreeFunction 769.1649
Outcome ~ Glucose+BloodPressure+SkinThickness+Insulin+BMI+Age 762.0105
Outcome ~ Glucose+BloodPressure+SkinThickness+Insulin+DiabetesPedigreeFunction+Age 792.9581
Outcome ~ Glucose+BloodPressure+SkinThickness+BMI+DiabetesPedigreeFunction+Age 754.9180
Outcome ~ Glucose+BloodPressure+Insulin+BMI+DiabetesPedigreeFunction+Age 752.6929
Outcome ~ Glucose+SkinThickness+Insulin+BMI+DiabetesPedigreeFunction+Age 758.2216
Outcome ~ BloodPressure+SkinThickness+Insulin+BMI+DiabetesPedigreeFunction+Age 865.8970
Outcome ~ Pregnancies+Glucose+BloodPressure+SkinThickness+Insulin+BMI+DiabetesPedigreeFunction 741.9676
Outcome ~ Pregnancies+Glucose+BloodPressure+SkinThickness+Insulin+BMI+Age 749.7849
Outcome ~ Pregnancies+Glucose+BloodPressure+SkinThickness+Insulin+DiabetesPedigreeFunction+Age 780.2247
Outcome ~ Pregnancies+Glucose+BloodPressure+SkinThickness+BMI+DiabetesPedigreeFunction+Age 741.1871
Outcome ~ Pregnancies+Glucose+BloodPressure+Insulin+BMI+DiabetesPedigreeFunction+Age 739.4534
Outcome ~ Pregnancies+Glucose+SkinThickness+Insulin+BMI+DiabetesPedigreeFunction+Age 745.9930
Outcome ~ Pregnancies+BloodPressure+SkinThickness+Insulin+BMI+DiabetesPedigreeFunction+Age 854.3724
Outcome ~ Glucose+BloodPressure+SkinThickness+Insulin+BMI+DiabetesPedigreeFunction+Age 754.6780
Outcome ~ Pregnancies+Glucose+BloodPressure+SkinThickness+Insulin+BMI+DiabetesPedigreeFunction+Age 741.4454

Ordena el dataframe por AIC de forma ascendente

# Ordena el dataframe por AIC de forma ascendente
modelos_aic <- modelos_aic[order (modelos_aic$AIC), ]

# Mostramos el dataframe
kable(modelos_aic, format = "html", escape = FALSE) %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed"),
                full_width = FALSE,
                position = "center",
                font_size = 12)
Formula AIC
251 Outcome ~ Pregnancies+Glucose+BloodPressure+Insulin+BMI+DiabetesPedigreeFunction+Age 739.4534
228 Outcome ~ Pregnancies+Glucose+BloodPressure+BMI+DiabetesPedigreeFunction+Age 739.4617
225 Outcome ~ Pregnancies+Glucose+BloodPressure+Insulin+BMI+DiabetesPedigreeFunction 739.9706
170 Outcome ~ Pregnancies+Glucose+BloodPressure+BMI+DiabetesPedigreeFunction 740.5596
250 Outcome ~ Pregnancies+Glucose+BloodPressure+SkinThickness+BMI+DiabetesPedigreeFunction+Age 741.1871
255 Outcome ~ Pregnancies+Glucose+BloodPressure+SkinThickness+Insulin+BMI+DiabetesPedigreeFunction+Age 741.4454
247 Outcome ~ Pregnancies+Glucose+BloodPressure+SkinThickness+Insulin+BMI+DiabetesPedigreeFunction 741.9676
222 Outcome ~ Pregnancies+Glucose+BloodPressure+SkinThickness+BMI+DiabetesPedigreeFunction 742.0039
179 Outcome ~ Pregnancies+Glucose+Insulin+BMI+DiabetesPedigreeFunction 743.5062
233 Outcome ~ Pregnancies+Glucose+Insulin+BMI+DiabetesPedigreeFunction+Age 744.1287
105 Outcome ~ Pregnancies+Glucose+BMI+DiabetesPedigreeFunction 744.3059
182 Outcome ~ Pregnancies+Glucose+BMI+DiabetesPedigreeFunction+Age 744.5088
176 Outcome ~ Pregnancies+Glucose+SkinThickness+BMI+DiabetesPedigreeFunction 745.0586
229 Outcome ~ Pregnancies+Glucose+SkinThickness+Insulin+BMI+DiabetesPedigreeFunction 745.3142
232 Outcome ~ Pregnancies+Glucose+SkinThickness+BMI+DiabetesPedigreeFunction+Age 745.5228
252 Outcome ~ Pregnancies+Glucose+SkinThickness+Insulin+BMI+DiabetesPedigreeFunction+Age 745.9930
171 Outcome ~ Pregnancies+Glucose+BloodPressure+BMI+Age 746.9861
226 Outcome ~ Pregnancies+Glucose+BloodPressure+Insulin+BMI+Age 747.9176
95 Outcome ~ Pregnancies+Glucose+BloodPressure+BMI 748.4285
167 Outcome ~ Pregnancies+Glucose+BloodPressure+Insulin+BMI 748.8900
223 Outcome ~ Pregnancies+Glucose+BloodPressure+SkinThickness+BMI+Age 748.9714
248 Outcome ~ Pregnancies+Glucose+BloodPressure+SkinThickness+Insulin+BMI+Age 749.7849
164 Outcome ~ Pregnancies+Glucose+BloodPressure+SkinThickness+BMI 750.2982
219 Outcome ~ Pregnancies+Glucose+BloodPressure+SkinThickness+Insulin+BMI 750.8479
106 Outcome ~ Pregnancies+Glucose+BMI+Age 752.0995
40 Outcome ~ Pregnancies+Glucose+BMI 752.1249
102 Outcome ~ Pregnancies+Glucose+Insulin+BMI 752.4279
244 Outcome ~ Glucose+BloodPressure+Insulin+BMI+DiabetesPedigreeFunction+Age 752.6929
180 Outcome ~ Pregnancies+Glucose+Insulin+BMI+Age 752.7533
207 Outcome ~ Glucose+BloodPressure+BMI+DiabetesPedigreeFunction+Age 753.2664
99 Outcome ~ Pregnancies+Glucose+SkinThickness+BMI 753.5999
177 Outcome ~ Pregnancies+Glucose+SkinThickness+BMI+Age 753.7527
173 Outcome ~ Pregnancies+Glucose+SkinThickness+Insulin+BMI 754.3985
254 Outcome ~ Glucose+BloodPressure+SkinThickness+Insulin+BMI+DiabetesPedigreeFunction+Age 754.6780
230 Outcome ~ Pregnancies+Glucose+SkinThickness+Insulin+BMI+Age 754.7446
243 Outcome ~ Glucose+BloodPressure+SkinThickness+BMI+DiabetesPedigreeFunction+Age 754.9180
212 Outcome ~ Glucose+Insulin+BMI+DiabetesPedigreeFunction+Age 756.2950
147 Outcome ~ Glucose+BMI+DiabetesPedigreeFunction+Age 757.2337
245 Outcome ~ Glucose+SkinThickness+Insulin+BMI+DiabetesPedigreeFunction+Age 758.2216
211 Outcome ~ Glucose+SkinThickness+BMI+DiabetesPedigreeFunction+Age 758.2431
136 Outcome ~ Glucose+BloodPressure+BMI+Age 759.5887
205 Outcome ~ Glucose+BloodPressure+Insulin+BMI+Age 760.1704
202 Outcome ~ Glucose+BloodPressure+SkinThickness+BMI+Age 761.5588
241 Outcome ~ Glucose+BloodPressure+SkinThickness+Insulin+BMI+Age 762.0105
71 Outcome ~ Glucose+BMI+Age 763.6846
145 Outcome ~ Glucose+Insulin+BMI+Age 763.9777
142 Outcome ~ Glucose+SkinThickness+BMI+Age 765.3455
209 Outcome ~ Glucose+SkinThickness+Insulin+BMI+Age 765.9777
204 Outcome ~ Glucose+BloodPressure+Insulin+BMI+DiabetesPedigreeFunction 767.2653
144 Outcome ~ Glucose+Insulin+BMI+DiabetesPedigreeFunction 767.5500
240 Outcome ~ Glucose+BloodPressure+SkinThickness+Insulin+BMI+DiabetesPedigreeFunction 769.1649
208 Outcome ~ Glucose+SkinThickness+Insulin+BMI+DiabetesPedigreeFunction 769.2725
135 Outcome ~ Glucose+BloodPressure+BMI+DiabetesPedigreeFunction 770.5636
141 Outcome ~ Glucose+SkinThickness+BMI+DiabetesPedigreeFunction 770.5938
201 Outcome ~ Glucose+BloodPressure+SkinThickness+BMI+DiabetesPedigreeFunction 770.8049
70 Outcome ~ Glucose+BMI+DiabetesPedigreeFunction 770.8709
132 Outcome ~ Glucose+BloodPressure+Insulin+BMI 775.4810
67 Outcome ~ Glucose+Insulin+BMI 775.7865
60 Outcome ~ Glucose+BloodPressure+BMI 777.0661
19 Outcome ~ Glucose+BMI 777.4030
198 Outcome ~ Glucose+BloodPressure+SkinThickness+Insulin+BMI 777.4792
138 Outcome ~ Glucose+SkinThickness+Insulin+BMI 777.7229
64 Outcome ~ Glucose+SkinThickness+BMI 778.2007
129 Outcome ~ Glucose+BloodPressure+SkinThickness+BMI 778.2337
41 Outcome ~ Pregnancies+Glucose+DiabetesPedigreeFunction 779.2744
174 Outcome ~ Pregnancies+Glucose+SkinThickness+Insulin+DiabetesPedigreeFunction 779.4930
220 Outcome ~ Pregnancies+Glucose+BloodPressure+SkinThickness+Insulin+DiabetesPedigreeFunction 779.5629
100 Outcome ~ Pregnancies+Glucose+SkinThickness+DiabetesPedigreeFunction 779.9315
165 Outcome ~ Pregnancies+Glucose+BloodPressure+SkinThickness+DiabetesPedigreeFunction 780.1796
96 Outcome ~ Pregnancies+Glucose+BloodPressure+DiabetesPedigreeFunction 780.2095
249 Outcome ~ Pregnancies+Glucose+BloodPressure+SkinThickness+Insulin+DiabetesPedigreeFunction+Age 780.2247
103 Outcome ~ Pregnancies+Glucose+Insulin+DiabetesPedigreeFunction 780.4989
107 Outcome ~ Pregnancies+Glucose+DiabetesPedigreeFunction+Age 780.6077
224 Outcome ~ Pregnancies+Glucose+BloodPressure+SkinThickness+DiabetesPedigreeFunction+Age 780.6210
231 Outcome ~ Pregnancies+Glucose+SkinThickness+Insulin+DiabetesPedigreeFunction+Age 780.7205
178 Outcome ~ Pregnancies+Glucose+SkinThickness+DiabetesPedigreeFunction+Age 780.9795
172 Outcome ~ Pregnancies+Glucose+BloodPressure+DiabetesPedigreeFunction+Age 781.2408
168 Outcome ~ Pregnancies+Glucose+BloodPressure+Insulin+DiabetesPedigreeFunction 781.5465
181 Outcome ~ Pregnancies+Glucose+Insulin+DiabetesPedigreeFunction+Age 781.9854
227 Outcome ~ Pregnancies+Glucose+BloodPressure+Insulin+DiabetesPedigreeFunction+Age 782.7657
38 Outcome ~ Pregnancies+Glucose+SkinThickness 789.9204
166 Outcome ~ Pregnancies+Glucose+BloodPressure+SkinThickness+Age 789.9463
93 Outcome ~ Pregnancies+Glucose+BloodPressure+SkinThickness 790.0628
163 Outcome ~ Pregnancies+Glucose+BloodPressure+SkinThickness+Insulin 790.0907
98 Outcome ~ Pregnancies+Glucose+SkinThickness+Insulin 790.1088
221 Outcome ~ Pregnancies+Glucose+BloodPressure+SkinThickness+Insulin+Age 790.1895
101 Outcome ~ Pregnancies+Glucose+SkinThickness+Age 790.5786
175 Outcome ~ Pregnancies+Glucose+SkinThickness+Insulin+Age 790.9478
9 Outcome ~ Pregnancies+Glucose 790.9500
37 Outcome ~ Pregnancies+Glucose+BloodPressure 792.0621
42 Outcome ~ Pregnancies+Glucose+Age 792.0834
39 Outcome ~ Pregnancies+Glucose+Insulin 792.7769
97 Outcome ~ Pregnancies+Glucose+BloodPressure+Age 792.8617
242 Outcome ~ Glucose+BloodPressure+SkinThickness+Insulin+DiabetesPedigreeFunction+Age 792.9581
210 Outcome ~ Glucose+SkinThickness+Insulin+DiabetesPedigreeFunction+Age 793.1387
72 Outcome ~ Glucose+DiabetesPedigreeFunction+Age 793.6857
94 Outcome ~ Pregnancies+Glucose+BloodPressure+Insulin 793.9369
203 Outcome ~ Glucose+BloodPressure+SkinThickness+DiabetesPedigreeFunction+Age 793.9574
104 Outcome ~ Pregnancies+Glucose+Insulin+Age 793.9880
143 Outcome ~ Glucose+SkinThickness+DiabetesPedigreeFunction+Age 794.0058
137 Outcome ~ Glucose+BloodPressure+DiabetesPedigreeFunction+Age 794.5402
169 Outcome ~ Pregnancies+Glucose+BloodPressure+Insulin+Age 794.8170
146 Outcome ~ Glucose+Insulin+DiabetesPedigreeFunction+Age 794.8213
206 Outcome ~ Glucose+BloodPressure+Insulin+DiabetesPedigreeFunction+Age 795.8361
200 Outcome ~ Glucose+BloodPressure+SkinThickness+Insulin+Age 801.3315
131 Outcome ~ Glucose+BloodPressure+SkinThickness+Age 801.5442
140 Outcome ~ Glucose+SkinThickness+Insulin+Age 801.8169
66 Outcome ~ Glucose+SkinThickness+Age 801.9037
139 Outcome ~ Glucose+SkinThickness+Insulin+DiabetesPedigreeFunction 802.2643
68 Outcome ~ Glucose+Insulin+DiabetesPedigreeFunction 802.8060
20 Outcome ~ Glucose+DiabetesPedigreeFunction 802.9874
21 Outcome ~ Glucose+Age 803.3621
199 Outcome ~ Glucose+BloodPressure+SkinThickness+Insulin+DiabetesPedigreeFunction 803.8132
62 Outcome ~ Glucose+BloodPressure+Age 804.2896
65 Outcome ~ Glucose+SkinThickness+DiabetesPedigreeFunction 804.4607
133 Outcome ~ Glucose+BloodPressure+Insulin+DiabetesPedigreeFunction 804.6953
61 Outcome ~ Glucose+BloodPressure+DiabetesPedigreeFunction 804.8254
69 Outcome ~ Glucose+Insulin+Age 805.1738
130 Outcome ~ Glucose+BloodPressure+SkinThickness+DiabetesPedigreeFunction 806.1477
134 Outcome ~ Glucose+BloodPressure+Insulin+Age 806.1720
63 Outcome ~ Glucose+SkinThickness+Insulin 811.8371
2 Outcome ~ Glucose 812.7196
17 Outcome ~ Glucose+SkinThickness 813.0668
128 Outcome ~ Glucose+BloodPressure+SkinThickness+Insulin 813.3188
18 Outcome ~ Glucose+Insulin 813.7694
16 Outcome ~ Glucose+BloodPressure 814.5946
58 Outcome ~ Glucose+BloodPressure+SkinThickness 814.6791
59 Outcome ~ Glucose+BloodPressure+Insulin 815.6765
253 Outcome ~ Pregnancies+BloodPressure+SkinThickness+Insulin+BMI+DiabetesPedigreeFunction+Age 854.3724
238 Outcome ~ Pregnancies+BloodPressure+Insulin+BMI+DiabetesPedigreeFunction+Age 854.8692
239 Outcome ~ Pregnancies+SkinThickness+Insulin+BMI+DiabetesPedigreeFunction+Age 856.2389
192 Outcome ~ Pregnancies+BloodPressure+BMI+DiabetesPedigreeFunction+Age 857.7617
197 Outcome ~ Pregnancies+Insulin+BMI+DiabetesPedigreeFunction+Age 857.8949
237 Outcome ~ Pregnancies+BloodPressure+SkinThickness+BMI+DiabetesPedigreeFunction+Age 859.4162
127 Outcome ~ Pregnancies+BMI+DiabetesPedigreeFunction+Age 860.2278
196 Outcome ~ Pregnancies+SkinThickness+BMI+DiabetesPedigreeFunction+Age 861.3496
190 Outcome ~ Pregnancies+BloodPressure+Insulin+BMI+Age 865.2376
235 Outcome ~ Pregnancies+BloodPressure+SkinThickness+Insulin+BMI+Age 865.6726
246 Outcome ~ BloodPressure+SkinThickness+Insulin+BMI+DiabetesPedigreeFunction+Age 865.8970
217 Outcome ~ BloodPressure+Insulin+BMI+DiabetesPedigreeFunction+Age 866.4310
218 Outcome ~ SkinThickness+Insulin+BMI+DiabetesPedigreeFunction+Age 866.9614
193 Outcome ~ Pregnancies+SkinThickness+Insulin+BMI+DiabetesPedigreeFunction 867.4738
194 Outcome ~ Pregnancies+SkinThickness+Insulin+BMI+Age 867.9545
234 Outcome ~ Pregnancies+BloodPressure+SkinThickness+Insulin+BMI+DiabetesPedigreeFunction 868.0758
157 Outcome ~ BloodPressure+BMI+DiabetesPedigreeFunction+Age 868.3643
125 Outcome ~ Pregnancies+Insulin+BMI+Age 868.4742
162 Outcome ~ Insulin+BMI+DiabetesPedigreeFunction+Age 868.4942
216 Outcome ~ BloodPressure+SkinThickness+BMI+DiabetesPedigreeFunction+Age 869.8990
92 Outcome ~ BMI+DiabetesPedigreeFunction+Age 870.0069
189 Outcome ~ Pregnancies+BloodPressure+Insulin+BMI+DiabetesPedigreeFunction 870.6060
116 Outcome ~ Pregnancies+BloodPressure+BMI+Age 870.7854
124 Outcome ~ Pregnancies+Insulin+BMI+DiabetesPedigreeFunction 870.8009
161 Outcome ~ SkinThickness+BMI+DiabetesPedigreeFunction+Age 871.0400
187 Outcome ~ Pregnancies+BloodPressure+SkinThickness+BMI+Age 872.7834
55 Outcome ~ Pregnancies+BMI+DiabetesPedigreeFunction 872.8374
115 Outcome ~ Pregnancies+BloodPressure+BMI+DiabetesPedigreeFunction 872.9567
121 Outcome ~ Pregnancies+SkinThickness+BMI+DiabetesPedigreeFunction 873.0775
56 Outcome ~ Pregnancies+BMI+Age 873.3333
186 Outcome ~ Pregnancies+BloodPressure+SkinThickness+BMI+DiabetesPedigreeFunction 873.6613
122 Outcome ~ Pregnancies+SkinThickness+BMI+Age 875.1755
155 Outcome ~ BloodPressure+Insulin+BMI+Age 875.4573
214 Outcome ~ BloodPressure+SkinThickness+Insulin+BMI+Age 875.8138
159 Outcome ~ SkinThickness+Insulin+BMI+Age 877.3072
90 Outcome ~ Insulin+BMI+Age 877.7959
81 Outcome ~ BloodPressure+BMI+Age 879.8930
118 Outcome ~ Pregnancies+SkinThickness+Insulin+BMI 880.5702
183 Outcome ~ Pregnancies+BloodPressure+SkinThickness+Insulin+BMI 881.0546
35 Outcome ~ BMI+Age 881.6754
152 Outcome ~ BloodPressure+SkinThickness+BMI+Age 881.8709
112 Outcome ~ Pregnancies+BloodPressure+Insulin+BMI 882.3473
52 Outcome ~ Pregnancies+Insulin+BMI 882.5455
87 Outcome ~ SkinThickness+BMI+Age 883.4654
13 Outcome ~ Pregnancies+BMI 887.1438
45 Outcome ~ Pregnancies+BloodPressure+BMI 887.3348
49 Outcome ~ Pregnancies+SkinThickness+BMI 888.5193
109 Outcome ~ Pregnancies+BloodPressure+SkinThickness+BMI 888.9702
158 Outcome ~ SkinThickness+Insulin+BMI+DiabetesPedigreeFunction 907.1214
213 Outcome ~ BloodPressure+SkinThickness+Insulin+BMI+DiabetesPedigreeFunction 909.1206
86 Outcome ~ SkinThickness+BMI+DiabetesPedigreeFunction 910.9413
89 Outcome ~ Insulin+BMI+DiabetesPedigreeFunction 912.3720
34 Outcome ~ BMI+DiabetesPedigreeFunction 912.5037
151 Outcome ~ BloodPressure+SkinThickness+BMI+DiabetesPedigreeFunction 912.9413
154 Outcome ~ BloodPressure+Insulin+BMI+DiabetesPedigreeFunction 914.2895
80 Outcome ~ BloodPressure+BMI+DiabetesPedigreeFunction 914.4484
126 Outcome ~ Pregnancies+Insulin+DiabetesPedigreeFunction+Age 915.6898
195 Outcome ~ Pregnancies+SkinThickness+Insulin+DiabetesPedigreeFunction+Age 917.1144
191 Outcome ~ Pregnancies+BloodPressure+Insulin+DiabetesPedigreeFunction+Age 917.5926
83 Outcome ~ SkinThickness+Insulin+BMI 918.7780
236 Outcome ~ Pregnancies+BloodPressure+SkinThickness+Insulin+DiabetesPedigreeFunction+Age 918.8826
148 Outcome ~ BloodPressure+SkinThickness+Insulin+BMI 920.7755
31 Outcome ~ Insulin+BMI 922.5638
123 Outcome ~ Pregnancies+SkinThickness+DiabetesPedigreeFunction+Age 922.7824
77 Outcome ~ BloodPressure+Insulin+BMI 924.4602
188 Outcome ~ Pregnancies+BloodPressure+SkinThickness+DiabetesPedigreeFunction+Age 924.5488
6 Outcome ~ BMI 924.7142
28 Outcome ~ SkinThickness+BMI 924.8473
57 Outcome ~ Pregnancies+DiabetesPedigreeFunction+Age 925.2260
24 Outcome ~ BloodPressure+BMI 926.6531
74 Outcome ~ BloodPressure+SkinThickness+BMI 926.8398
91 Outcome ~ Insulin+DiabetesPedigreeFunction+Age 926.8503
117 Outcome ~ Pregnancies+BloodPressure+DiabetesPedigreeFunction+Age 927.2254
53 Outcome ~ Pregnancies+Insulin+DiabetesPedigreeFunction 927.4275
160 Outcome ~ SkinThickness+Insulin+DiabetesPedigreeFunction+Age 928.1656
156 Outcome ~ BloodPressure+Insulin+DiabetesPedigreeFunction+Age 928.8018
113 Outcome ~ Pregnancies+BloodPressure+Insulin+DiabetesPedigreeFunction 929.2218
119 Outcome ~ Pregnancies+SkinThickness+Insulin+DiabetesPedigreeFunction 929.2752
215 Outcome ~ BloodPressure+SkinThickness+Insulin+DiabetesPedigreeFunction+Age 930.0006
184 Outcome ~ Pregnancies+BloodPressure+SkinThickness+Insulin+DiabetesPedigreeFunction 931.1239
54 Outcome ~ Pregnancies+Insulin+Age 931.7618
120 Outcome ~ Pregnancies+SkinThickness+Insulin+Age 932.0934
88 Outcome ~ SkinThickness+DiabetesPedigreeFunction+Age 932.7015
114 Outcome ~ Pregnancies+BloodPressure+Insulin+Age 933.7002
185 Outcome ~ Pregnancies+BloodPressure+SkinThickness+Insulin+Age 933.8083
153 Outcome ~ BloodPressure+SkinThickness+DiabetesPedigreeFunction+Age 934.5315
36 Outcome ~ DiabetesPedigreeFunction+Age 935.1037
50 Outcome ~ Pregnancies+SkinThickness+DiabetesPedigreeFunction 935.4908
14 Outcome ~ Pregnancies+DiabetesPedigreeFunction 936.4855
82 Outcome ~ BloodPressure+DiabetesPedigreeFunction+Age 937.0948
110 Outcome ~ Pregnancies+BloodPressure+SkinThickness+DiabetesPedigreeFunction 937.3233
46 Outcome ~ Pregnancies+BloodPressure+DiabetesPedigreeFunction 937.9182
51 Outcome ~ Pregnancies+SkinThickness+Age 940.2193
33 Outcome ~ Insulin+Age 941.0204
85 Outcome ~ SkinThickness+Insulin+Age 941.2639
111 Outcome ~ Pregnancies+BloodPressure+SkinThickness+Age 941.9259
79 Outcome ~ BloodPressure+Insulin+Age 942.9899
150 Outcome ~ BloodPressure+SkinThickness+Insulin+Age 943.0419
12 Outcome ~ Pregnancies+Insulin 945.3401
48 Outcome ~ Pregnancies+SkinThickness+Insulin 946.4545
44 Outcome ~ Pregnancies+BloodPressure+Insulin 946.9929
15 Outcome ~ Pregnancies+Age 947.0981
30 Outcome ~ SkinThickness+Age 948.1391
108 Outcome ~ Pregnancies+BloodPressure+SkinThickness+Insulin 948.2765
47 Outcome ~ Pregnancies+BloodPressure+Age 949.0640
76 Outcome ~ BloodPressure+SkinThickness+Age 949.9084
8 Outcome ~ Age 954.7203
11 Outcome ~ Pregnancies+SkinThickness 955.3627
26 Outcome ~ BloodPressure+Age 956.6631
43 Outcome ~ Pregnancies+BloodPressure+SkinThickness 957.1700
1 Outcome ~ Pregnancies 960.2099
10 Outcome ~ Pregnancies+BloodPressure 961.2553
78 Outcome ~ BloodPressure+Insulin+DiabetesPedigreeFunction 968.9094
32 Outcome ~ Insulin+DiabetesPedigreeFunction 969.0193
149 Outcome ~ BloodPressure+SkinThickness+Insulin+DiabetesPedigreeFunction 970.8758
84 Outcome ~ SkinThickness+Insulin+DiabetesPedigreeFunction 971.0133
25 Outcome ~ BloodPressure+DiabetesPedigreeFunction 974.1017
7 Outcome ~ DiabetesPedigreeFunction 974.8609
75 Outcome ~ BloodPressure+SkinThickness+DiabetesPedigreeFunction 975.2200
29 Outcome ~ SkinThickness+DiabetesPedigreeFunction 975.3182
23 Outcome ~ BloodPressure+Insulin 984.4124
5 Outcome ~ Insulin 984.8104
73 Outcome ~ BloodPressure+SkinThickness+Insulin 986.3133
27 Outcome ~ SkinThickness+Insulin 986.4636
22 Outcome ~ BloodPressure+SkinThickness 993.0831
4 Outcome ~ SkinThickness 993.1890
3 Outcome ~ BloodPressure 994.1276

Resumir el modelo final

#Resumir el modelo final

resumen_final <- tidy (modelo_final, conf.int = TRUE) %>%
  filter (term != "(Intercept) ") %>%
  mutate (odds.ratio = round (exp (estimate) , 2),
          CI_lower = round (exp (conf.low), 2),
          CI_upper = round (exp (conf.high), 2) ) %>%
  select (term, odds.ratio, CI_lower, CI_upper, p.value)

#Aplicar formato condicional a los valores p

bold_if_significant <- function (p_value) {
  p_value_rounded <- round (p_value, 4)
  if (p_value < 0.05) {
    cell_spec (p_value_rounded, format = "html", bold = TRUE, color = "#8B0AB0")
    }
  else{
    as.character (p_value_rounded)
  }
}

resumen_final$p.value <- sapply (resumen_final$p.value, bold_if_significant)

# Imprimir el resumen del modelo final usando la variable fórmula
cat ("El Modelo Final se compone de las siguiente variables:", formula_mejor_modelo, ": \n")
El Modelo Final se compone de las siguiente variables: Outcome ~ Pregnancies+Glucose+BloodPressure+Insulin+BMI+DiabetesPedigreeFunction+Age : 
kable(resumen_final, format = "html", escape = FALSE) %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed"),
                full_width = FALSE,
                position = "center",
                font_size = 12)
term odds.ratio CI_lower CI_upper p.value
(Intercept) 4469.97 1151.02 19165.57 0
Pregnancies 0.88 0.83 0.94 1e-04
Glucose 0.97 0.96 0.97 0
BloodPressure 1.01 1.00 1.02 0.0103
Insulin 1.00 1.00 1.00 0.1553
BMI 0.91 0.89 0.94 0
DiabetesPedigreeFunction 0.39 0.21 0.69 0.0015
Age 0.99 0.97 1.00 0.1114

Volver al inicio