-
Notifications
You must be signed in to change notification settings - Fork 1
/
R_Logistische_Regression.R
214 lines (149 loc) · 7.71 KB
/
R_Logistische_Regression.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
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
#######################################
# Logistische Regression:
# Datensatz 1 - nur signifikante Variablen
final_data_relevant_1_v4 <- final_data_relevant_1_v4[,c("FRAUD_1_NONFRAUD_0", "DSRI", "PMR_4", "DEF_TAX_EXP_VJ",
"ZSCORE_4_J", "ZSCORE_4_VJ", "POS_NEG_O_CF_J", "POS_NEG_O_CF_VJ",
"J_var3_VJ", "MJ_var2_J", "MJ_var2_4", "entity_specific_tag", "issue_sum")]
colnames(final_data_relevant_1_v4)
final_data_relevant_1_v4$FRAUD_1_NONFRAUD_0 <- as.factor(final_data_relevant_1_v4$FRAUD_1_NONFRAUD_0)
final_data_relevant_1_v4$POS_NEG_O_CF_J <- as.factor(final_data_relevant_1_v4$POS_NEG_O_CF_J)
final_data_relevant_1_v4$POS_NEG_O_CF_VJ <- as.factor(final_data_relevant_1_v4$POS_NEG_O_CF_VJ)
final_data_relevant_1_v4$issue_sum <- as.factor(final_data_relevant_1_v4$issue_sum)
str(final_data_relevant_1_v4)
final_data_relevant_1_v4$FRAUD_1_NONFRAUD_0 <- as.factor(final_data_relevant_1_v4$FRAUD_1_NONFRAUD_0)
set.seed(123)
sens_results<-c()
spec_results<-c()
results_test<-c()
sens_results_train<-c()
spec_results_train<-c()
results_train<-c()
#?glm
Grenzwert <- 0.5 # Eingabe Grenzwert
for (i in 1:1000) {
training.samples <- final_data_relevant_1_v4$FRAUD_1_NONFRAUD_0 %>%
createDataPartition(p = 0.7, list = FALSE)
# Zuweisung der Indexwerte zu Train und Test
train.data <- final_data_relevant_1_v4[training.samples, ]
test.data <- final_data_relevant_1_v4[-training.samples, ]
both <- ovun.sample(FRAUD_1_NONFRAUD_0~., data=train.data, method = "over", p= 0.5,seed=1234)$data
# Fit the model
model <- glm( FRAUD_1_NONFRAUD_0 ~ .,
data = both, family = binomial)
predicted_train <- predict(model, train.data, type="response")
predicted_test <- predict(model, test.data, type="response")
round(predicted_test, 2)
summary(model)
predicted_result<-ifelse(predicted_test > Grenzwert, 1, 0)
predicted_result_train<-ifelse(predicted_train > Grenzwert, 1, 0)
result<-mean(predicted_result == test.data$FRAUD_1_NONFRAUD_0)
results_test<-c(results_test, result)
result_train<-mean(predicted_result_train == train.data$FRAUD_1_NONFRAUD_0)
results_train<-c(results_train, result_train)
# Ermittlung des Spezifität und Sensitivität TEST
actual_values<-test.data$FRAUD_1_NONFRAUD_0
conf_matrix<-table(predicted_result,actual_values)
sens<-conf_matrix[4]/(conf_matrix[3]+conf_matrix[4])
spec<-conf_matrix[1]/(conf_matrix[1]+conf_matrix[2])
# Spezifität, Sensitivität TRAIN
train_values<-train.data$FRAUD_1_NONFRAUD_0
conf_matrix_tr<-table(predicted_result_train,train_values)
sens_train<-conf_matrix_tr[4]/(conf_matrix_tr[3]+conf_matrix_tr[4])
spec_train<-conf_matrix_tr[1]/(conf_matrix_tr[1]+conf_matrix_tr[2])
# Einzelergebnisse in die Vektoren einfügen
sens_results<-c(sens_results, sens)
spec_results<-c(spec_results, spec)
sens_results_train<-c(sens_results_train, sens_train)
spec_results_train<-c(spec_results_train, spec_train)
}
sens_results[is.na(sens_results)]<-0
spec_results[is.na(spec_results)]<-0
sens_results_train[is.na(sens_results_train)]<-0
spec_results_train[is.na(spec_results_train)]<-0
mean(sens_results)*100
mean(spec_results)*100
mean(results_test)*100
mean(sens_results_train)*100
mean(spec_results_train)*100
mean(results_train)*100
round(predicted_test, 2)
summary(model)
# Datensatz 1 - nicht signifikante Variablen
final_data_relevant_1_v2<-final_data_relevant_1[, names(final_data_relevant_1) %in%
c("FRAUD_1_NONFRAUD_0",
"entity_specific_tag", "entity_specific_tag_VJ",
"issue_sum", "POS_NEG_O_CF_J", "POS_NEG_O_CF_VJ",
"POS_NEG_O_CF_4", "PMR_J", "PMR_VJ", "PMR_4",
"ACCR_CF_J", "ACCR_CF_VJ", "ACCR_CF_4",
"DEF_TAX_EXP_J", "DEF_TAX_EXP_VJ",
"J_var3_J", "J_var3_VJ", "J_var3_4",
"F_Score_Result",
"MJ_var2_J", "MJ_var2_VJ", "MJ_var2_4",
"DSRI", "DSRI_1",
"ABW_CF_ABS_MJ",
"ZSCORE_4_J", "ZSCORE_4_VJ"
)]
colnames(final_data_relevant_1_v2)
final_data_relevant_1_v2$FRAUD_1_NONFRAUD_0 <- as.factor(final_data_relevant_1_v2$FRAUD_1_NONFRAUD_0)
final_data_relevant_1_v2$POS_NEG_O_CF_J <- as.factor(final_data_relevant_1_v2$POS_NEG_O_CF_J)
final_data_relevant_1_v2$POS_NEG_O_CF_VJ <- as.factor(final_data_relevant_1_v2$POS_NEG_O_CF_VJ)
final_data_relevant_1_v2$issue_sum <- as.factor(final_data_relevant_1_v2$issue_sum)
str(final_data_relevant_1_v2)
final_data_relevant_1_v2$FRAUD_1_NONFRAUD_0 <- as.factor(final_data_relevant_1_v2$FRAUD_1_NONFRAUD_0)
set.seed(123)
sens_results<-c()
spec_results<-c()
results_test<-c()
sens_results_train<-c()
spec_results_train<-c()
results_train<-c()
#?glm
Grenzwert <- 0.5
for (i in 1:1000) {
training.samples <- final_data_relevant_1_v2$FRAUD_1_NONFRAUD_0 %>%
createDataPartition(p = 0.7, list = FALSE)
# Zuweisung der Indexwerte zu Train und Test
train.data <- final_data_relevant_1_v2[training.samples, ]
test.data <- final_data_relevant_1_v2[-training.samples, ]
both <- ovun.sample(FRAUD_1_NONFRAUD_0~., data=train.data, method = "over", p= 0.5,seed=1234)$data
# Fit the model
model <- glm( FRAUD_1_NONFRAUD_0 ~ .,
data = both, family = binomial)
predicted_train <- predict(model, train.data, type="response")
predicted_test <- predict(model, test.data, type="response")
round(predicted_test, 2)
summary(model)
predicted_result<-ifelse(predicted_test > Grenzwert, 1, 0)
predicted_result_train<-ifelse(predicted_train > Grenzwert, 1, 0)
result<-mean(predicted_result == test.data$FRAUD_1_NONFRAUD_0)
results_test<-c(results_test, result)
result_train<-mean(predicted_result_train == train.data$FRAUD_1_NONFRAUD_0)
results_train<-c(results_train, result_train)
# Ermittlung des Spezifität und Sensitivität TEST
actual_values<-test.data$FRAUD_1_NONFRAUD_0
conf_matrix<-table(predicted_result,actual_values)
sens<-conf_matrix[4]/(conf_matrix[3]+conf_matrix[4])
spec<-conf_matrix[1]/(conf_matrix[1]+conf_matrix[2])
# Spezifität, Sensitivität TRAIN
train_values<-train.data$FRAUD_1_NONFRAUD_0
conf_matrix_tr<-table(predicted_result_train,train_values)
sens_train<-conf_matrix_tr[4]/(conf_matrix_tr[3]+conf_matrix_tr[4])
spec_train<-conf_matrix_tr[1]/(conf_matrix_tr[1]+conf_matrix_tr[2])
# Einzelergebnisse in die Vektoren einfügen
sens_results<-c(sens_results, sens)
spec_results<-c(spec_results, spec)
sens_results_train<-c(sens_results_train, sens_train)
spec_results_train<-c(spec_results_train, spec_train)
}
sens_results[is.na(sens_results)]<-0
spec_results[is.na(spec_results)]<-0
sens_results_train[is.na(sens_results_train)]<-0
spec_results_train[is.na(spec_results_train)]<-0
mean(sens_results)*100
mean(spec_results)*100
mean(results_test)*100
mean(sens_results_train)*100
mean(spec_results_train)*100
mean(results_train)*100
round(predicted_test, 2)
summary(model)