-
Notifications
You must be signed in to change notification settings - Fork 5
/
Copy path.Rhistory
232 lines (232 loc) · 9.31 KB
/
.Rhistory
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
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
# Load necessary libraries
library(dplyr)
library(ggplot2)
# Load the data
data <- read.csv("/Users/shangjunyao/DATA550/midterm_project/covid_sub.csv")
# Check for missing values
summary(data)
# Check the percentage of missing values for each column
sapply(data, function(x) mean(is.na(x)) * 100)
num_columns <- ncol(data)
print(num_columns)
dimensions <- dim(data)
num_columns <- dimensions[2] # The second element is the number of columns
print(num_columns)
rm dimensions
setwd("/Users/shangjunyao/DATA550/midterm_project")
# Define the main project directory name
project_dir <- "Covid-19-Data-Analysis"
# Create the project directory
dir.create(project_dir)
# Create subdirectories within the main directory
dir.create(file.path(project_dir, "data"))
dir.create(file.path(project_dir, "scripts"))
dir.create(file.path(project_dir, "scripts", "demographics_analysis"))
dir.create(file.path(project_dir, "scripts", "comorbidity_analysis"))
dir.create(file.path(project_dir, "scripts", "severity_treatment_analysis"))
dir.create(file.path(project_dir, "scripts", "mortality_analysis"))
dir.create(file.path(project_dir, "output"))
dir.create(file.path(project_dir, "config"))
dir.create(file.path(project_dir, "report"))
file.copy("/Users/shangjunyao/DATA550/midterm_project/covid_sub.csv", file.path(project_dir, "data"))
file.copy("/Users/shangjunyao/DATA550/midterm_project/covid_sub.csv", file.path(project_dir, "data"))
file.create(file.path(project_dir, "config", "config.yaml"))
file.create(file.path(project_dir, "README.md"))
# Set working directory to your midterm project folder
setwd("/Users/shangjunyao/DATA550/midterm_project")
# Create the main folder structure
dir.create("data")
dir.create("scripts")
dir.create("scripts/demographics_analysis")
dir.create("scripts/comorbidity_analysis")
dir.create("scripts/severity_treatment_analysis")
dir.create("scripts/mortality_analysis")
dir.create("output")
dir.create("config")
dir.create("report")
file.copy("/Users/shangjunyao/DATA550/midterm_project/covid_sub.csv", file.path(project_dir, "data"))
file.copy("/Users/shangjunyao/DATA550/midterm_project/covid_sub.csv", "data/covid_sub.csv")
file.create("config.yaml")
file.create("config.yaml")
file.create("README.md")
file.create("config/config.yaml")
data <- read.csv("/Users/shangjunyao/DATA550/midterm_project/covid_sub.csv")
colSums(is.na(data))
date_died_column <- data$date_died
# Remove DATE_DIED temporarily from the dataset
data_without_date_died <- data[, !names(data) %in% "date_died"]
# Apply na.omit() to remove rows with missing values in other columns
data_cleaned <- na.omit(data_without_date_died)
# Add back the DATE_DIED column (only for rows that remain after na.omit)
data_cleaned$date_died <- date_died_column[rownames(data_cleaned)]
View(data_cleaned)
colSums(is.na(data_cleaned))
data_cleaned <- na.omit(data[, !names(data) %in% "date_died"])
# Re-add DATE_DIED column from the original dataset
data_cleaned$date_died <- data$date_died[rownames(data_cleaned)]
# Verify that only DATE_DIED may have missing values
colSums(is.na(data_cleaned))
data <- read.csv("/Users/shangjunyao/DATA550/midterm_project/covid_sub.csv")
data_cleaned <- na.omit(data[, !names(data) %in% "date_died"])
# Re-add DATE_DIED column from the original dataset
data_cleaned$date_died <- data$date_died[rownames(data_cleaned)]
# Verify that only DATE_DIED may have missing values
colSums(is.na(data_cleaned))
data <- read.csv("/Users/shangjunyao/DATA550/midterm_project/covid_sub.csv")
rows_to_keep <- complete.cases(data[, !names(data) %in% "date_died"])
# Subset the data to include only rows without missing values in other columns
data_cleaned <- data[rows_to_keep, ]
# Verify that only DATE_DIED may still contain missing values
colSums(is.na(data_cleaned))
View(data_cleaned)
View(data)
View(data_cleaned)
data <- read.csv("/Users/shangjunyao/DATA550/midterm_project/covid_sub.csv")
data_clean <- data %>%
filter(if_all(SEX ~ !is.na(.x)))%>%
filter(if_all(INTUBED ~ !is.na(.x)))%>%
filter(if_all(PNEUMONIA ~ !is.na(.x))) %>%
filter(if_all(AGE ~ !is.na(.x)))%>%
filter(if_all(PREGNANT ~ !is.na(.x)))%>%
filter(if_all(DIABETES ~ !is.na(.x)))%>%
filter(if_all(COPD ~ !is.na(.x)))%>%
filter(if_all(ASTHMA ~ !is.na(.x)))%>%
filter(if_all(INMSUPR ~ !is.na(.x)))%>%
filter(if_all(HYPERYENSION ~ !is.na(.x)))%>%
filter(if_all(OTHER_DISEASE ~ !is.na(.x)))%>%
filter(if_all(CARDIOVASCULAR ~ !is.na(.x)))%>%
filter(if_all(OBESITY ~ !is.na(.x)))%>%
filter(if_all(RENAL_CHRONIC ~ !is.na(.x)))%>%
filter(if_all(TOBACCO ~ !is.na(.x)))%>%
filter(if_all(ICU ~ !is.na(.x)))%>%
data_clean <- data %>%
filter(if_all(SEX ~ !is.na(.x)))%>%
filter(if_all(INTUBED ~ !is.na(.x)))%>%
filter(if_all(PNEUMONIA ~ !is.na(.x))) %>%
filter(if_all(AGE ~ !is.na(.x)))%>%
filter(if_all(PREGNANT ~ !is.na(.x)))%>%
filter(if_all(DIABETES ~ !is.na(.x)))%>%
filter(if_all(COPD ~ !is.na(.x)))%>%
filter(if_all(ASTHMA ~ !is.na(.x)))%>%
filter(if_all(INMSUPR ~ !is.na(.x)))%>%
filter(if_all(HYPERYENSION ~ !is.na(.x)))%>%
filter(if_all(OTHER_DISEASE ~ !is.na(.x)))%>%
filter(if_all(CARDIOVASCULAR ~ !is.na(.x)))%>%
filter(if_all(OBESITY ~ !is.na(.x)))%>%
filter(if_all(RENAL_CHRONIC ~ !is.na(.x)))%>%
filter(if_all(TOBACCO ~ !is.na(.x)))%>%
filter(if_all(ICU ~ !is.na(.x)))
columns_to_filter <- c("SEX", "INTUBED", "PNEUMONIA", "AGE", "PREGNANT", "DIABETES",
"COPD", "ASTHMA", "INMSUPR", "HYPERTENSION", "OTHER_DISEASE",
"CARDIOVASCULAR", "OBESITY", "RENAL_CHRONIC", "TOBACCO", "ICU")
# Apply filters to remove rows with missing values in the specified columns
data_clean <- data %>%
filter(if_all(all_of(columns_to_filter), ~ !is.na(.)))
# Define the columns to filter for missing values, excluding DATE_DIED
columns_to_filter <- c("SEX", "INTUBED", "PNEUMONIA", "AGE", "PREGNANT", "DIABETES",
"COPD", "ASTHMA", "INMSUPR", "HIPERTENSION", "OTHER_DISEASE",
"CARDIOVASCULAR", "OBESITY", "RENAL_CHRONIC", "TOBACCO", "ICU")
# Apply filters to remove rows with missing values in the specified columns
data_clean <- data %>%
filter(if_all(all_of(columns_to_filter), ~ !is.na(.)))
View(data_clean)
colSums(is.na(data_clean))
library(dplyr)
data <- read.csv("/Users/shangjunyao/DATA550/midterm_project/covid_sub.csv")
#Check the missing data
colSums(is.na(data))
#Clean the missing data except date_died
columns_to_filter <- c("SEX", "INTUBED", "PNEUMONIA", "AGE", "PREGNANT", "DIABETES",
"COPD", "ASTHMA", "INMSUPR", "HIPERTENSION", "OTHER_DISEASE",
"CARDIOVASCULAR", "OBESITY", "RENAL_CHRONIC", "TOBACCO", "ICU")
data_clean <- data %>%
filter(if_all(all_of(columns_to_filter), ~ !is.na(.)))
# Check the result
colSums(is.na(data_clean))
write.csv(data_clean, "/Users/shangjunyao/DATA550/midterm_project/data/data_clean.csv", row.names=FALSE)
install.packages("haven")
install.packages("survival")
install.packages("survminer")
# Load libraries
library(haven)
library(survival)
library(survminer)
# Load the dataset
vets <- read_sas("/Users/shangjunyao/Desktop/vets.sas7bdat")
setwd /Users/shangjunyao/Desktop/
vets$STATUS <- factor(vets$STATUS, levels = c(0, 1), labels = c("Censored", "Died"))
log_TX <- survfit(Surv(SURVT,STATUS) ~ TX, data=vets )
ggsurvplot(survfit(Surv(SURVT, STATUS) ~ TX, data = vets), fun = "cloglog", title = "Log-Log Survival Curves for TX")
ggsurvplot(survfit(Surv(SURVT, STATUS) ~ TX, data = vets),
fun = "cloglog")
cox_tx <- coxph(Surv(SURVT, STATUS == "Died") ~ TX, data = vets)
test_ph_tx <- cox.zph(cox_tx)
cox_tx <- coxph(Surv(SURVT, STATUS == "Died") ~ TX, data = vets)
test_ph_tx <- cox.zph(cox_tx)
cox_tx <- coxph(Surv(SURVT, STATUS) ~ TX, data = vets, ties="breslow")
cox.zph(cox_tx)
coxph(Surv(SURVT, STATUS)~TX+tt(TX),
data = vet,
ties = "breslow",
tt = function(x,t,...) as.numeric(x)*t
)
coxph(Surv(SURVT, STATUS)~TX+tt(TX),
data = vets,
ties = "breslow",
tt = function(x,t,...) as.numeric(x)*t
)
coxph(Surv(SURVT, STATUS)~TX+tt(TX),
data = vets,
ties = "breslow",
tt = function(x,t,...) as.numeric(x)*log(t)
)
ggsurvplot(survfit(Surv(SURVT, STATUS) ~ AGE, data = vets),
fun = "cloglog")
View(vets)
# Define age groups: less than 65, 65-80, and greater than 80
vets$AGEGROUP <- cut(vets$AGE,
breaks = c(-Inf, 65, 80, Inf),
labels = c("Less than 65", "65-80", "Greater than 80"))
# Verify the distribution of age groups
table(vets$AGEGROUP)
median(vets$AGE)
vets$age_group2<-ifelse(vets$AGE <62, "Less than 62", "older than 62")
ggsurvplot(survfit(Surv(SURVT, STATUS) ~ age_group2, data = vets),
fun = "cloglog")
vets$age_group3 <- cut(vets$AGE,
breaks = 3,
labels = c(1, 2, 3),
include.lowest = TRUE)
ggsurvplot(survfit(Surv(SURVT, STATUS) ~ age_group3, data = vets),
fun = "cloglog")
cox_age <- coxph(Surv(SURVT, STATUS) ~ AGE, data = vets, ties="breslow")
cox.zph(cox_age)
coxph(Surv(SURVT, STATUS)~AGE+tt(AGE),
data = vets,
ties = "breslow",
tt = function(x,t,...) as.numeric(x)*t
)
coxph(Surv(SURVT, STATUS)~AGE+tt(AGE),
data = vets,
ties = "breslow",
tt = function(x,t,...) as.numeric(x)*log(t)
)
ggsurvplot(survfit(Surv(SURVT, STATUS) ~ PRIORTX, data = vets),
fun = "cloglog")
cox_p <- coxph(Surv(SURVT, STATUS) ~ PRIORTX, data = vets, ties="breslow")
cox.zph(cox_p)
coxph(Surv(SURVT, STATUS)~PRIORTX+tt(PRIORTX),
data = vets,
ties = "breslow",
tt = function(x,t,...) as.numeric(x)*t
)
coxph(Surv(SURVT, STATUS)~PRIORTX+tt(PRIORTX),
data = vets,
ties = "breslow",
tt = function(x,t,...) as.numeric(x)*log(t)
)
library(survival)
# Fit the Cox Proportional Hazards model controlling for AGE
cox_model_prior_age <- coxph(Surv(SURVT, STATUS) ~ PRIORTX + AGE, data = vets)
# Display the results
summary(cox_model_prior_age)