-
Notifications
You must be signed in to change notification settings - Fork 0
/
OR_table_new.R
156 lines (110 loc) · 4.24 KB
/
OR_table_new.R
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
#=================================================
# File: Build OR(95% Ci) Table
# Author: Joy Fu
# Update Date: 07/24/2019
#=================================================
#=================================================
# Quick Reference
# Function to build OR(95% CI) table:
# extract_OR = function(regression_object, variable_index_lst, mark)
# make_OR_table = function(lst_model, var_position, model_head)
# Function to build coefficient table:
# extract_coeff = function(regression_object, variable_index_lst, mark)
# make_coeff_table = function(lst_model, model_head, type = c('normal', 'interact'), var_position = NULL)
#=================================================
extract_OR = function(regression_object, variable_index_lst, mark) {
summary_reg = summary(regression_object)
coefficient = summary_reg$coefficients
or_table = array(NA, dim=c(length(variable_index_lst),8))
for(i in 1:length(variable_index_lst)) {
index = variable_index_lst[i]
OR = exp(coefficient[index])
confinterval = exp(confint(regression_object, level = 0.95))
num_variable = length(confinterval)/2
lower_CI = confinterval[index]
upper_CI = confinterval[num_variable + index]
CI_95 = paste0('(', sprintf('%.2f',lower_CI), ', ', sprintf('%.2f',upper_CI), ')')
p = round(coefficient[num_variable*3 + index], 4)
or_table[i,1] = mark
or_table[i,2] = index
or_table[i,3] = length(regression_object$y)
or_table[i,4] = round(OR, 2)
or_table[i,5] = CI_95
or_table[i,6] = sprintf('%.3f',p)
or_table[i,7] = round(lower_CI, 2)
or_table[i,8] = round(upper_CI, 2)
i = i + 1
}
or_table_final = as.data.frame(or_table)
colnames(or_table_final) = c('mark','index', 'N', 'OR', '95% CI', 'p-value', 'lower_CI', 'upper_CI')
return(or_table_final)
}
make_OR_table = function(lst_model, var_position, model_head) {
# Make to a full table
or_table = data.frame(
Mark = character(),
Index = integer(),
N = integer(),
OR = double(),
CI_95 = character(),
p_value = character(),
lower_CI = double(),
upper_CI = double(),
stringsAsFactors = F
)
for (i in 1:length(lst_model)) {
model_object = get(lst_model[i])
result = extract_OR(model_object, var_position, paste0(model_head, lst_model[i]))
or_table = rbind(or_table, result)
i = i + 1
}
return(or_table)
}
#=====================================================
#
# Additional: Extract the coefficients for regression
#
#=====================================================
extract_coeff = function(regression_object, variable_index_lst, mark) {
summary_reg = summary(regression_object)
coefficient = summary_reg$coefficients
coef_table = array(NA, dim=c(length(variable_index_lst),5))
for(i in 1:length(variable_index_lst)) {
index = variable_index_lst[i]
coeff = coefficient[index]
confinterval = exp(confint(regression_object, level = 0.95))
num_variable = length(confinterval)/2
p = round(coefficient[num_variable*3 + index], 4)
coef_table[i,1] = mark
coef_table[i,2] = index
coef_table[i,3] = length(regression_object$y)
coef_table[i,4] = sprintf('%.3f',coeff)
coef_table[i,5] = sprintf('%.3f',p)
i = i + 1
}
coef_table_final = as.data.frame(coef_table)
colnames(coef_table_final) = c('mark','index', 'N', 'coeff', 'p-value')
return(coef_table_final)
}
make_coeff_table = function(lst_model, model_head, type = c('normal', 'interact'), var_position = NULL) {
# Make to a full table
coeff_table = data.frame(
Mark = character(),
Index = integer(),
N = integer(),
coeff = double(),
p_value = double(),
stringsAsFactors = F
)
for (i in 1:length(lst_model)) {
model_object = get(lst_model[i])
try = summary(model_object)
if (type == "interact") {
var_position = length(try$coefficients)/4
}
result = extract_coeff(model_object, var_position, paste0(model_head, lst_model[i]))
coeff_table = rbind(coeff_table, result)
i = i + 1
}
return(coeff_table)
}