Nube de datos: rep

Mostrando entradas con la etiqueta rep. Mostrar todas las entradas

2019-12-28

How to convert a continuous variable to discrete in R?

Problem

We want to convert continuous variable to discrete in R:

'Create a new qualitative variable, called Elite, by binning the Top10perc variable. We are going to divide universities into two groups based on whether or not the proportion of students coming from the top 10% of their high school classes exceeds 50%'.

library(ISLR)
library(tidyverse)
glimpse(College)

Observations: 777
Variables: 18
$ Private      Yes, Yes, Yes, Yes, Yes, Yes, Yes, Yes, Yes, Yes, Yes, Ye...
$ Apps         1660, 2186, 1428, 417, 193, 587, 353, 1899, 1038, 582, 17...
$ Accept       1232, 1924, 1097, 349, 146, 479, 340, 1720, 839, 498, 142...
$ Enroll       721, 512, 336, 137, 55, 158, 103, 489, 227, 172, 472, 484...
$ Top10perc    23, 16, 22, 60, 16, 38, 17, 37, 30, 21, 37, 44, 38, 44, 2...
$ Top25perc    52, 29, 50, 89, 44, 62, 45, 68, 63, 44, 75, 77, 64, 73, 4...
$ F.Undergrad  2885, 2683, 1036, 510, 249, 678, 416, 1594, 973, 799, 183...
$ P.Undergrad  537, 1227, 99, 63, 869, 41, 230, 32, 306, 78, 110, 44, 63...
$ Outstate     7440, 12280, 11250, 12960, 7560, 13500, 13290, 13868, 155...
$ Room.Board   3300, 6450, 3750, 5450, 4120, 3335, 5720, 4826, 4400, 338...
$ Books        450, 750, 400, 450, 800, 500, 500, 450, 300, 660, 500, 40...
$ Personal     2200, 1500, 1165, 875, 1500, 675, 1500, 850, 500, 1800, 6...
$ PhD          70, 29, 53, 92, 76, 67, 90, 89, 79, 40, 82, 73, 60, 79, 3...
$ Terminal     78, 30, 66, 97, 72, 73, 93, 100, 84, 41, 88, 91, 84, 87, ...
$ S.F.Ratio    18.1, 12.2, 12.9, 7.7, 11.9, 9.4, 11.5, 13.7, 11.3, 11.5,...
$ perc.alumni  12, 16, 30, 37, 2, 11, 26, 37, 23, 15, 31, 41, 21, 32, 26...
$ Expend       7041, 10527, 8735, 19016, 10922, 9727, 8861, 11487, 11644...
$ Grad.Rate    60, 56, 54, 59, 15, 55, 63, 73, 80, 52, 73, 76, 74, 68, 5...

Solution

Option 1: form ISLR's book.

Elite = rep("No", nrow(College))
Elite[College$Top10perc > 50] = "Yes"
Elite <- as.factor(Elite)
college <- data.frame(College,  Elite)
summary(college[, c("Top10perc", "Elite")])

  Top10perc     Elite    
 Min.   : 1.00   No :699  
 1st Qu.:15.00   Yes: 78  
 Median :23.00            
 Mean   :27.56            
 3rd Qu.:35.00            
 Max.   :96.00

Option 2: ifelse from base package and dplyr

# base 
College$Elite <- factor(ifelse(College$Top10perc > 50, "Yes", "No"))
# dplyr
library(dplyr)
College <-
  college %>%
  mutate(Elite = factor(ifelse(College$Top10perc > 50, "Yes", "No")))

Option 3: creating a logical vector.

There are multiple options. I show two examples.

college$Elite <- transform(College, Elite = Top10perc > 50)
College$Elite <- College$Top10perc > 50

References

An Introduction to Statistical Learning - Page 54.

From 'An Introduction to Statistical Learning' (ISLR), page 54.

2018-12-08

Discretización de variables en R

Problema

Deseamos discretizar una variable, es decir, convertir una variable continua en discreta. Utilizamos el conjunto de datos College del paquete ISLR. Crearemos una nueva variable cualitativa llamada Elite, discretizando la variable Top10perc. Vamos a dividir las universidades en dos grupos basados en si la proporción de nuevos estudiantes provienen de entre el 10% de los mejores alumnos de sus institutos excede o no el 50%.

library(ISLR)
library(tidyverse)
glimpse(College)

Observations: 777
Variables: 18
$ Private      Yes, Yes, Yes, Yes, Yes, Yes, Yes, Yes, Yes, Yes, Yes, Ye...
$ Apps         1660, 2186, 1428, 417, 193, 587, 353, 1899, 1038, 582, 17...
$ Accept       1232, 1924, 1097, 349, 146, 479, 340, 1720, 839, 498, 142...
$ Enroll       721, 512, 336, 137, 55, 158, 103, 489, 227, 172, 472, 484...
$ Top10perc    23, 16, 22, 60, 16, 38, 17, 37, 30, 21, 37, 44, 38, 44, 2...
$ Top25perc    52, 29, 50, 89, 44, 62, 45, 68, 63, 44, 75, 77, 64, 73, 4...
$ F.Undergrad  2885, 2683, 1036, 510, 249, 678, 416, 1594, 973, 799, 183...
$ P.Undergrad  537, 1227, 99, 63, 869, 41, 230, 32, 306, 78, 110, 44, 63...
$ Outstate     7440, 12280, 11250, 12960, 7560, 13500, 13290, 13868, 155...
$ Room.Board   3300, 6450, 3750, 5450, 4120, 3335, 5720, 4826, 4400, 338...
$ Books        450, 750, 400, 450, 800, 500, 500, 450, 300, 660, 500, 40...
$ Personal     2200, 1500, 1165, 875, 1500, 675, 1500, 850, 500, 1800, 6...
$ PhD          70, 29, 53, 92, 76, 67, 90, 89, 79, 40, 82, 73, 60, 79, 3...
$ Terminal     78, 30, 66, 97, 72, 73, 93, 100, 84, 41, 88, 91, 84, 87, ...
$ S.F.Ratio    18.1, 12.2, 12.9, 7.7, 11.9, 9.4, 11.5, 13.7, 11.3, 11.5,...
$ perc.alumni  12, 16, 30, 37, 2, 11, 26, 37, 23, 15, 31, 41, 21, 32, 26...
$ Expend       7041, 10527, 8735, 19016, 10922, 9727, 8861, 11487, 11644...
$ Grad.Rate    60, 56, 54, 59, 15, 55, 63, 73, 80, 52, 73, 76, 74, 68, 5...

Solución

Opción 1:Propuesta en el libro ISLR.

Elite = rep("No", nrow(College))
Elite[College$Top10perc > 50] = "Yes"
Elite <- as.factor(Elite)
college <- data.frame(College,  Elite)
summary(college[, c("Top10perc", "Elite")])

  Top10perc     Elite    
 Min.   : 1.00   No :699  
 1st Qu.:15.00   Yes: 78  
 Median :23.00            
 Mean   :27.56            
 3rd Qu.:35.00            
 Max.   :96.00

Opción 2: ifelse con paquete base y dplyr

# base 
College$Elite <- factor(ifelse(College$Top10perc > 50, "Yes", "No"))
# dplyr
library(dplyr)
College <-
  college %>%
  mutate(Elite = factor(ifelse(College$Top10perc > 50, "Yes", "No")))

Opción 3: vector lógico.

Hay múltiples opciones. Presento dos ejemplos.

college$Elite <- transform(College, Elite = Top10perc > 50)
College$Elite <- College$Top10perc > 50

Entradas relacionadas

2018-01-20

Rellenar área con teselas en ggplot2

Problema

Con ggplot2 queremos rellenar una región cuadrada con teselas cuadradas. El intento original del usuario que plantéo la pregunta fue:

Datos

set.seed(1)
library(ggplot2)
# Datos
l = 1000
a = seq(0, 1, 1 / (l - 1))
x = rep(a, each = length(a))
y = rep(a, length(a))
k = length(x)
c = sample(1:10, k, replace = TRUE)
data <- data.frame(x, y, c)
# Gráfico
ggplot(data, aes(x = x, y = y)) + geom_point(shape = 15, color = c)

Solución

Alternativa 1

ggplot(data, aes(x = x, y = y, fill = c)) + geom_tile(colour = "white")

Alternativa 2

colors<-c("peachpuff", "yellow", "orange", "orangered", "red", 
          "darkred","firebrick", "royalblue", "darkslategrey", "black")
ggplot(data, aes(x = x, y = y)) +
  geom_tile(aes(fill = factor(c)), colour = "white") +
  scale_fill_manual(values = colors, name = "Colours") +
  coord_equal()

Notas

Para apreciar lo que sucede cuando creamos el gráfico original con geom_point, reducimos el tamaño del data frame, a 10 x 10. Lo representamos seguido de las dos alternativas propuestas.

Nuevo data frame

l = 100
a = seq(0, 1, 1 / (l - 1))
x = rep(a, each = length(a))
y = rep(a, length(a))
k = length(x)
c = sample(1:10, k, replace = TRUE)
data <- data.frame(x, y, c)

Gráfico original

ggplot(data, aes(x = x, y = y)) + geom_point(shape = 15, color = c)

Alternativa 1

ggplot(data, aes(x = x, y = y, fill = c)) + geom_tile(colour = "white")

Alternativa 2

colors<-c("peachpuff", "yellow", "orange", "orangered", "red", 
          "darkred","firebrick", "royalblue", "darkslategrey", "black")
ggplot(data, aes(x = x, y = y)) +
  geom_tile(aes(fill = factor(c)), colour = "white") +
  scale_fill_manual(values = colors, name = "Colours") +
  coord_equal()

Nube de datos

2019-12-28

How to convert a continuous variable to discrete in R?

Problem

Solution

References

Related posts

2018-12-08

Discretización de variables en R

Problema

Solución

Entradas relacionadas

2018-01-20

Rellenar área con teselas en ggplot2

Problema

Solución

Notas

Referencias