-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathMCA.Rmd
131 lines (115 loc) · 5.16 KB
/
MCA.Rmd
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
---
title: "MCA report"
output: github_document
---
```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
```
#### load libraries
```{r include=FALSE}
library(readr)
library(tidyverse)
library(dplyr)
library(ggplot2)
library(FactoMineR)
library(factoextra)
```
#### load data
``` {r}
data <- read_csv("data.csv")
summary(data)
str(data)
names(data)<-str_replace_all(names(data), c(" " = "." , "," = "" ))
head(data)
```
#### data manipulation
``` {r}
data$BMI.Category <- gsub("Normal Weight", "Normal", data$BMI.Category)
data$BMI.Category <- factor(data$BMI.Category, levels = c("Normal","Overweight","Obese"))
data$sleep.hours <- as.factor(ifelse(data$Sleep.Duration < 7, '< 7 hours',
ifelse(data$Sleep.Duration > 7, '7 - 9 hours', '7 hours')))
data <- data %>%
mutate(Stress.Level = as.character(Stress.Level)) %>%
mutate(Quality.of.Sleep = as.character(Quality.of.Sleep))
data$Stress.Level <- factor(data$Stress.Level,levels = c("1","2","3","4","5","6","7","8","9","10"))
data$Quality.of.Sleep <- factor(data$Quality.of.Sleep,levels = c("1","2","3","4","5","6","7","8","9","10"))
data$agecat <- as.factor(ifelse(data$Age < 31, '20 - 30',
ifelse(data$Age < 41, '30 - 40',
ifelse(data$Age < 51, '40 - 50', '50 - 60'))))
data$bloodpressure <- as.factor(ifelse(data$Blood.Pressure %in% c("117/76","118/76","115/75","115/78","119/77","118/75"), 'normal',
ifelse(data$Blood.Pressure %in% c("121/79"), 'elevated',
ifelse(data$Blood.Pressure %in% c("120/80","125/80","122/80","126/83","132/87","130/86","128/85","131/86","128/84","135/88","129/84","130/85","125/82"), 'high bp stage 1', 'high bp stage 2'))))
data$bloodpressure <- factor(data$bloodpressure, levels = c("normal", "elevated", "high bp stage 1", "high bp stage 2"))
data$dailysteps <- as.factor(ifelse(data$Daily.Steps < 5000, 'sedentary',
ifelse(data$Daily.Steps > 4999 & data$Daily.Steps < 7500, 'low active',
ifelse(data$Daily.Steps > 7499 & data$Daily.Steps < 10000, 'somewhat active', 'active'))))
data$dailysteps <- factor(data$dailysteps, levels = c("sedentary","low active","somewhat active","active"))
data$heart.rate <- case_when(
data$Gender == "Male" & data$Heart.Rate > 70 & data$Heart.Rate < 76 ~ "average",
data$Gender == "Male" & data$Heart.Rate > 65 & data$Heart.Rate < 72 ~ "above average",
data$Gender == "Male" & data$Heart.Rate > 74 & data$Heart.Rate < 81 ~ "below average",
data$Gender == "Male" & data$Heart.Rate > 54 & data$Heart.Rate < 62 ~ "excellent",
data$Gender == "Male" & data$Heart.Rate > 61 & data$Heart.Rate < 67 ~ "good",
data$Gender == "Male" & data$Heart.Rate > 81 ~ "poor",
data$Gender == "Female" & data$Heart.Rate > 72 & data$Heart.Rate < 77 ~ "average",
data$Gender == "Female" & data$Heart.Rate > 68 & data$Heart.Rate < 74 ~ "above average",
data$Gender == "Female" & data$Heart.Rate > 76 & data$Heart.Rate < 84 ~ "below average",
data$Gender == "Female" & data$Heart.Rate > 59 & data$Heart.Rate < 65 ~ "excellent",
data$Gender == "Female" & data$Heart.Rate > 64 & data$Heart.Rate < 69 ~ "good",
data$Gender == "Female" & data$Heart.Rate > 83 ~ "poor")
data$heart.rate <- factor(data$heart.rate, levels = c("poor", "below average", "average", "above average","good","excellent"))
data$Sleep.Disorder <- factor(data$Sleep.Disorder, levels = c("None", "Insomnia", "Sleep Apnea"))
str(data)
```
#### subset
``` {r}
dataMCA <- subset(data, select = c(2,4,6,8,9,13:18))
head(dataMCA)
```
#### mca
``` {r}
res.mca <- MCA(dataMCA, quali.sup = c(1,2,8), graph = FALSE)
summary(res.mca)
head(res.mca$eig, 10)
fviz_screeplot(res.mca, addlabels = TRUE, ylim = c(0, 45))
```
#### dimdesc
``` {r}
res.desc <- dimdesc(res.mca, axes = c(1,2))
# Description of dimension 1
res.desc[[1]]
# Description of dimension 2
res.desc[[2]]
```
#### categories variables
``` {r}
var <- get_mca_var(res.mca)
var
fviz_mca_var(res.mca, choice = "mca.cor",
repel = TRUE, ggtheme = theme_minimal(),
geom = c("arrow", "text"), arrow.size = 0.02, arrow.length = unit(0.1, "inches"))
#coord var
var$coord
fviz_mca_var(res.mca, col.var="black", shape.var = 15,
repel = TRUE,ggtheme = theme_minimal(),linecolor = "black")
head(round(var$coord, 2),15)
```
#### cos2
``` {r}
head(round(var$cos2, 2),15)
fviz_mca_var(res.mca, col.var = "cos2", alpha.var="cos2",invisible=c("quali.sup","ind"),
gradient.cols = c("#00AFBB", "#E7B800", "#FC4E07"),
repel = TRUE, # Avoid text overlapping
ggtheme = theme_minimal())
library("corrplot")
corrplot(var$cos2, is.corr=FALSE,method = "number",
tl.col = "brown",tl.cex = 0.8, tl.srt=45, number.cex = 0.5)
```
#### contrib
``` {r}
head(round(var$contrib, 2),15)
fviz_mca_var(res.mca, col.var = "contrib", invisible=c("quali.sup","ind"),
gradient.cols = c("#00AFBB", "#E7B800", "#FC4E07"),
repel = TRUE, # avoid text overlapping (slow)
ggtheme = theme_minimal())
```