Library:
> # install.packages("knitr")
> # install.packages("rmarkdown")
> # install.packages("prettydoc")
> # install.packages("equatiomatic")1 PENDAHULUAN
1.1 Latar Belakang
Regresi logistik adalah salah satu metode analisis statistik yang digunakan untuk memodelkan hubungan variabel independent terhadap variabel dependent yang bersakala data nominal/ordinal.
2 TINJAUAN PUSTAKA
2.1 Regresi Logistik
Regresi logistik akan menghasilkan dua kemungkinan nilai, seperti “berhasil/gagal”, “ya/tidak”, atau “1/0”. Tujuan utama regresi logistik adalah untuk memprediksi probabilitas kejadian suatu peristiwa berdasarkan variabel-variabel prediktor yang ada. Metode analisis regresi logistik menggunakan fungsi logit atau log-odds ratio untuk menghubungkan variabel prediktor dengan probabilitas kejadian peristiwa yang dimodelkan. Fungsi logit ini mengubah nilai probabilitas menjadi nilai logaritma odds ratio, yang menggambarkan hubungan linier antara prediktor dan variabel dependen biner.
2.2 Asumsi Regresi Logistik
- Independensi : Asumsi independensi menyatakan bahwa pengamatan dalam sampel adalah independen satu sama lain. Variabel independen tidak memerlukan asumsi multivariate normality.
Regresi logistik tidak membutuhkan hubungan linier antara variabel independen dengan variabel dependen (Ghozali, Imam. 2013).
- Asumsi Multikoliniertias Untuk mengetahui korelasi antara dua variabel bebas. Pengujian menggunakan nilai variance inflation factor (VIF). Asumsi multikolinieritas terpenuhi apabila nilai VIF < 10.
2.3 Uji Signifikansi keseluruhan Model
Pengujian regresi logistik secara serentak dilakukan untuk mengetahui apakah model telah signifikan dengan hipotesis sebagai berikut.
Hipotesis 𝐻0∶𝛽1=𝛽2=⋯=𝛽𝑝=0 𝐻1∶ minimal ada satu 𝛽𝑗≠0,𝑗=1,2,…,𝑝
Keputusan Tolak H0 jika nilai statistik uji-G nilai 𝐺 > Xhitung atau p-value statistik uji-G < alpha (0,05).
2.4 Odds Ratio
Odds merupakan rasio probabilitas sukses (π) terhadap probabilitas gagal ( 1-π). Nilai odds bernilai positif (0 < odds ∞ ) Odds ratio dilambangkan dengan psi (𝜓).
Rumus 𝜓=exp(𝛽jk)
3 SOURCE CODE
Berikut ini merupakan tahapan melakukan analisis regresi logistik pada Rstudio:
3.1 Library
> library(readr)
> library(generalhoslem)
> library(pscl)
> library(car)3.2 Input Data
Selanjutnya, kita dapat memuat data ke dalam Rstudio.
> # Input data yang akan dimuat
> data<-read.csv("C:/Users/Salma/Downloads/heartcleveland.csv")
> str(data)
'data.frame': 297 obs. of 14 variables:
$ age : int 69 69 66 65 64 64 63 61 60 59 ...
$ sex : int 1 0 0 1 1 1 1 1 0 1 ...
$ cp : int 0 0 0 0 0 0 0 0 0 0 ...
$ trestbps : int 160 140 150 138 110 170 145 134 150 178 ...
$ chol : int 234 239 226 282 211 227 233 234 240 270 ...
$ fbs : int 1 0 0 1 0 0 1 0 0 0 ...
$ restecg : int 2 0 0 2 2 2 2 0 0 2 ...
$ thalach : int 131 151 114 174 144 155 150 145 171 145 ...
$ exang : int 0 0 0 0 1 0 0 0 0 0 ...
$ oldpeak : num 0.1 1.8 2.6 1.4 1.8 0.6 2.3 2.6 0.9 4.2 ...
$ slope : int 1 0 2 1 1 1 2 1 0 2 ...
$ ca : int 1 2 0 1 0 0 0 2 0 0 ...
$ thal : int 0 0 0 0 0 2 1 0 0 2 ...
$ condition: int 0 0 0 1 0 0 0 1 0 0 ...
> Y <- as.factor(data$oldpeak)
> X1 <- data$cp
> X2 <- as.factor(data$fbs)
> X3 <- as.factor(data$condition)
> str(Y)
Factor w/ 40 levels "0","0.1","0.2",..: 2 18 26 15 18 7 23 26 10 37 ...
>
> #Membentuk data frame
> data_logistik<-data.frame(X1,X2,X3,Y)
> str(data_logistik)
'data.frame': 297 obs. of 4 variables:
$ X1: int 0 0 0 0 0 0 0 0 0 0 ...
$ X2: Factor w/ 2 levels "0","1": 2 1 1 2 1 1 2 1 1 1 ...
$ X3: Factor w/ 2 levels "0","1": 1 1 1 2 1 1 1 2 1 1 ...
$ Y : Factor w/ 40 levels "0","0.1","0.2",..: 2 18 26 15 18 7 23 26 10 37 ...3.3 Asumsi Nonmultikolinieritas
> #Asumsi Nonmultikolinieritas
> reg1 <- lm(X1~X2+X3, data=data_logistik)
> summary(reg1)
Call:
lm(formula = X1 ~ X2 + X3, data = data_logistik)
Residuals:
Min 1Q Median 3Q Max
-2.6075 -0.6075 0.3444 0.3925 1.3444
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.81695 0.07277 24.970 < 2e-16 ***
X21 -0.16139 0.14538 -1.110 0.268
X31 0.79055 0.10262 7.703 2.04e-13 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Residual standard error: 0.8816 on 294 degrees of freedom
Multiple R-squared: 0.1707, Adjusted R-squared: 0.1651
F-statistic: 30.26 on 2 and 294 DF, p-value: 1.121e-12
>
> reglog2 <- glm(X2~X1+X3, family = binomial, data =data_logistik)
> summary(reglog2)
Call:
glm(formula = X2 ~ X1 + X3, family = binomial, data = data_logistik)
Coefficients:
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.4414 0.3742 -3.852 0.000117 ***
X1 -0.1979 0.1783 -1.110 0.267097
X31 0.1766 0.3616 0.489 0.625186
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
(Dispersion parameter for binomial family taken to be 1)
Null deviance: 245.65 on 296 degrees of freedom
Residual deviance: 244.45 on 294 degrees of freedom
AIC: 250.45
Number of Fisher Scoring iterations: 4
> pR2(reglog2)
fitting null model for pseudo-r2
llh llhNull G2 McFadden r2ML
-1.222252e+02 -1.228239e+02 1.197544e+00 4.875045e-03 4.024018e-03
r2CU
7.151482e-03
>
> reglog3 <- glm(X3~X1+X2, family = binomial, data =data_logistik)
> summary(reglog3)
Call:
glm(formula = X3 ~ X1 + X2, family = binomial, data = data_logistik)
Coefficients:
Estimate Std. Error z value Pr(>|z|)
(Intercept) -2.5349 0.4077 -6.217 5.07e-10 ***
X1 1.0486 0.1615 6.495 8.30e-11 ***
X21 0.1892 0.3646 0.519 0.604
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
(Dispersion parameter for binomial family taken to be 1)
Null deviance: 409.95 on 296 degrees of freedom
Residual deviance: 354.66 on 294 degrees of freedom
AIC: 360.66
Number of Fisher Scoring iterations: 4
> pR2(reglog3)
fitting null model for pseudo-r2
llh llhNull G2 McFadden r2ML r2CU
-177.3287662 -204.9732479 55.2889634 0.1348687 0.1698577 0.2269324 3.4 Analisis Regresi Logistik
> reglog<-glm(Y~X1+X2+X3,family=binomial,data=data_logistik)
> summary(reglog)
Call:
glm(formula = Y ~ X1 + X2 + X3, family = binomial, data = data_logistik)
Coefficients:
Estimate Std. Error z value Pr(>|z|)
(Intercept) 0.02219 0.30900 0.072 0.942760
X1 0.10604 0.14286 0.742 0.457932
X21 0.27843 0.37965 0.733 0.463331
X31 1.11999 0.29089 3.850 0.000118 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
(Dispersion parameter for binomial family taken to be 1)
Null deviance: 373.79 on 296 degrees of freedom
Residual deviance: 351.42 on 293 degrees of freedom
AIC: 359.42
Number of Fisher Scoring iterations: 43.5 Uji Signifikansi Keseluruhan Model
> pR2(reglog)
fitting null model for pseudo-r2
llh llhNull G2 McFadden r2ML
-175.71081868 -186.89673232 22.37182728 0.05985077 0.07255893
r2CU
0.10134801
> qchisq(0.95,2)
[1] 5.9914653.6 Uji Parsial dan R Square
> #Uji Parsial Parameter Model
> summary(reglog)
Call:
glm(formula = Y ~ X1 + X2 + X3, family = binomial, data = data_logistik)
Coefficients:
Estimate Std. Error z value Pr(>|z|)
(Intercept) 0.02219 0.30900 0.072 0.942760
X1 0.10604 0.14286 0.742 0.457932
X21 0.27843 0.37965 0.733 0.463331
X31 1.11999 0.29089 3.850 0.000118 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
(Dispersion parameter for binomial family taken to be 1)
Null deviance: 373.79 on 296 degrees of freedom
Residual deviance: 351.42 on 293 degrees of freedom
AIC: 359.42
Number of Fisher Scoring iterations: 4
>
> #R square
> summary(reglog)
Call:
glm(formula = Y ~ X1 + X2 + X3, family = binomial, data = data_logistik)
Coefficients:
Estimate Std. Error z value Pr(>|z|)
(Intercept) 0.02219 0.30900 0.072 0.942760
X1 0.10604 0.14286 0.742 0.457932
X21 0.27843 0.37965 0.733 0.463331
X31 1.11999 0.29089 3.850 0.000118 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
(Dispersion parameter for binomial family taken to be 1)
Null deviance: 373.79 on 296 degrees of freedom
Residual deviance: 351.42 on 293 degrees of freedom
AIC: 359.42
Number of Fisher Scoring iterations: 4
> Rsq<-1-(351.42/373.79)
> Rsq
[1] 0.059846443.7 Asumsi Multikolinieritas
> vif(reglog)
X1 X2 X3
1.164896 1.006824 1.157597 3.8 Odds Ratio
> beta<-(coef(reglog))
> beta
(Intercept) X1 X21 X31
0.02218662 0.10604036 0.27842765 1.11998562
> OR_beta<-exp(beta)
> OR_beta
(Intercept) X1 X21 X31
1.022435 1.111867 1.321051 3.064810
> cbind(beta,OR_beta)
beta OR_beta
(Intercept) 0.02218662 1.022435
X1 0.10604036 1.111867
X21 0.27842765 1.321051
X31 1.11998562 3.0648103.9 Membentuk klasifikasi
> yp_hat<-fitted(reglog)
> data_logistik$yp_hat<-yp_hat
> data_logistik
X1 X2 X3 Y yp_hat
1 0 1 0 0.1 0.5745927
2 0 0 0 1.8 0.5055464
3 0 0 0 2.6 0.5055464
4 0 1 1 1.4 0.8054324
5 0 0 0 1.8 0.5055464
6 0 0 0 0.6 0.5055464
7 0 1 0 2.3 0.5745927
8 0 0 1 2.6 0.7580782
9 0 0 0 0.9 0.5055464
10 0 0 0 4.2 0.5055464
11 0 0 1 0.2 0.7580782
12 0 0 1 0 0.7580782
13 0 0 1 0.8 0.7580782
14 0 1 0 1 0.5745927
15 0 0 0 1.9 0.5055464
16 0 0 0 0 0.5055464
17 0 1 0 1.2 0.5745927
18 0 0 0 1.4 0.5055464
19 0 0 1 1.2 0.7580782
20 0 0 0 0.8 0.5055464
21 0 0 0 1.4 0.5055464
22 0 0 1 3.8 0.7580782
23 0 0 0 0 0.5055464
24 1 0 0 0.2 0.5320129
25 1 0 0 0.4 0.5320129
26 1 0 0 0 0.5320129
27 1 0 1 0 0.7769903
28 1 0 0 0 0.5320129
29 1 0 1 1.4 0.7769903
30 1 1 0 0 0.6002855
31 1 0 0 0 0.5320129
32 1 0 1 1.8 0.7769903
33 1 1 1 0 0.8215142
34 1 0 1 0 0.7769903
35 1 0 1 0.3 0.7769903
36 1 0 1 0 0.7769903
37 1 0 0 0 0.5320129
38 1 0 0 1.3 0.5320129
39 1 0 0 0.8 0.5320129
40 1 0 0 0 0.5320129
41 1 0 0 1.4 0.5320129
42 1 0 0 0 0.5320129
43 1 0 0 1.2 0.5320129
44 1 0 0 0 0.5320129
45 1 1 0 0 0.6002855
46 1 0 1 0 0.7769903
47 1 0 0 0.2 0.5320129
48 1 0 0 0.8 0.5320129
49 1 1 0 0 0.6002855
50 1 0 0 1.1 0.5320129
51 1 0 0 0 0.5320129
52 1 0 0 0.6 0.5320129
53 1 0 1 1 0.7769903
54 1 0 0 0.2 0.5320129
55 1 1 0 0 0.6002855
56 1 0 0 0 0.5320129
57 1 0 0 0.6 0.5320129
58 1 0 0 0 0.5320129
59 1 0 0 0 0.5320129
60 1 0 0 0 0.5320129
61 1 0 0 0 0.5320129
62 1 0 0 0 0.5320129
63 1 0 0 0 0.5320129
64 1 0 0 0 0.5320129
65 1 0 0 1.4 0.5320129
66 1 0 0 0 0.5320129
67 1 0 0 0 0.5320129
68 1 0 0 0 0.5320129
69 1 0 0 0 0.5320129
70 1 0 0 0 0.5320129
71 1 0 0 0.7 0.5320129
72 1 0 0 0 0.5320129
73 2 0 0 1.1 0.5583004
74 2 1 0 0 0.6254380
75 2 0 1 2.9 0.7948241
76 2 0 1 2 0.7948241
77 2 1 1 1.6 0.8365365
78 2 0 0 1.5 0.5583004
79 2 0 0 1 0.5583004
80 2 0 0 1.6 0.5583004
81 2 0 1 0.8 0.7948241
82 2 0 0 0 0.5583004
83 2 0 0 0 0.5583004
84 2 1 0 0.8 0.6254380
85 2 0 0 0.8 0.5583004
86 2 0 0 0.8 0.5583004
87 2 0 1 1.8 0.7948241
88 2 0 1 0 0.7948241
89 2 0 0 0.2 0.5583004
90 2 0 0 0 0.5583004
91 2 0 0 1.8 0.5583004
92 2 0 1 1.2 0.7948241
93 2 1 0 1 0.6254380
94 2 0 1 3 0.7948241
95 2 0 0 0 0.5583004
96 2 1 0 0 0.6254380
97 2 1 1 2.2 0.8365365
98 2 1 0 1.6 0.6254380
99 2 0 1 2.5 0.7948241
100 2 0 0 0.6 0.5583004
101 2 0 1 3.2 0.7948241
102 2 0 0 0 0.5583004
103 2 1 0 0 0.6254380
104 2 0 1 0.4 0.7948241
105 2 0 0 1.6 0.5583004
106 2 1 0 0.2 0.6254380
107 2 1 1 0.6 0.8365365
108 2 0 0 0.5 0.5583004
109 2 0 0 0.4 0.5583004
110 2 0 0 1.6 0.5583004
111 2 1 0 0 0.6254380
112 2 0 0 1.6 0.5583004
113 2 0 0 0 0.5583004
114 2 0 0 0 0.5583004
115 2 1 0 1.2 0.6254380
116 2 1 0 0 0.6254380
117 2 0 0 0.1 0.5583004
118 2 1 0 0.5 0.6254380
119 2 0 0 1.2 0.5583004
120 2 1 0 2.4 0.6254380
121 2 0 0 0.6 0.5583004
122 2 0 0 1.5 0.5583004
123 2 0 0 0.5 0.5583004
124 2 0 0 0 0.5583004
125 2 0 0 0.6 0.5583004
126 2 0 0 1.6 0.5583004
127 2 0 1 0.6 0.7948241
128 2 0 0 0 0.5583004
129 2 0 1 2 0.7948241
130 2 0 1 0.8 0.7948241
131 2 0 0 0.2 0.5583004
132 2 1 0 0 0.6254380
133 2 0 0 0 0.5583004
134 2 0 1 0 0.7948241
135 2 0 0 0 0.5583004
136 2 0 0 1.4 0.5583004
137 2 0 1 3.6 0.7948241
138 2 0 0 0.6 0.5583004
139 2 0 0 0.3 0.5583004
140 2 0 0 0.4 0.5583004
141 2 0 0 0 0.5583004
142 2 0 0 0 0.5583004
143 2 0 0 0.2 0.5583004
144 2 0 0 1.9 0.5583004
145 2 1 0 0.8 0.6254380
146 2 0 0 0 0.5583004
147 2 0 0 0 0.5583004
148 2 0 0 2 0.5583004
149 2 0 0 0 0.5583004
150 2 0 0 0 0.5583004
151 2 0 0 0 0.5583004
152 2 0 0 0 0.5583004
153 2 0 0 0 0.5583004
154 2 0 0 3.5 0.5583004
155 2 0 0 0 0.5583004
156 3 0 1 0 0.8115775
157 3 0 0 1.6 0.5842653
158 3 0 1 2.6 0.8115775
159 3 0 1 2.4 0.8115775
160 3 1 1 3.4 0.8505245
161 3 0 1 1.5 0.8115775
162 3 0 1 2.6 0.8115775
163 3 1 1 0.2 0.8505245
164 3 0 1 0.9 0.8115775
165 3 0 1 1 0.8115775
166 3 0 0 0.3 0.5842653
167 3 0 0 0.4 0.5842653
168 3 1 1 1 0.8505245
169 3 0 1 0.1 0.8115775
170 3 0 0 2.3 0.5842653
171 3 0 1 1 0.8115775
172 3 0 1 2.8 0.8115775
173 3 0 0 0.4 0.5842653
174 3 0 1 0.6 0.8115775
175 3 0 1 2.2 0.8115775
176 3 0 1 2 0.8115775
177 3 0 0 2 0.5842653
178 3 0 0 0.2 0.5842653
179 3 0 0 0 0.5842653
180 3 0 1 1.4 0.8115775
181 3 0 1 4 0.8115775
182 3 0 1 1.8 0.8115775
183 3 0 1 0 0.8115775
184 3 1 1 1.8 0.8505245
185 3 0 1 4 0.8115775
186 3 0 1 3.6 0.8115775
187 3 0 1 6.2 0.8115775
188 3 0 1 1.8 0.8115775
189 3 0 0 1.2 0.5842653
190 3 1 1 1.9 0.8505245
191 3 0 1 1.4 0.8115775
192 3 0 0 0 0.5842653
193 3 0 1 3.6 0.8115775
194 3 0 1 1 0.8115775
195 3 0 1 3.6 0.8115775
196 3 0 1 0 0.8115775
197 3 0 1 1.9 0.8115775
198 3 0 1 0 0.8115775
199 3 0 1 2.4 0.8115775
200 3 0 1 2.8 0.8115775
201 3 0 1 2.8 0.8115775
202 3 0 1 2.6 0.8115775
203 3 0 1 1.2 0.8115775
204 3 1 1 1.4 0.8505245
205 3 0 1 1.4 0.8115775
206 3 0 1 0 0.8115775
207 3 0 1 3.4 0.8115775
208 3 0 0 0.5 0.5842653
209 3 0 1 1.2 0.8115775
210 3 0 1 0 0.8115775
211 3 1 1 1 0.8505245
212 3 0 1 0 0.8115775
213 3 0 0 0 0.5842653
214 3 0 1 4.4 0.8115775
215 3 0 1 2.2 0.8115775
216 3 0 0 1 0.5842653
217 3 0 1 3 0.8115775
218 3 0 1 2 0.8115775
219 3 0 0 0.6 0.5842653
220 3 1 1 2.8 0.8505245
221 3 0 1 0.8 0.8115775
222 3 0 1 0 0.8115775
223 3 0 1 0.1 0.8115775
224 3 0 0 0.4 0.5842653
225 3 0 1 0.6 0.8115775
226 3 1 1 1 0.8505245
227 3 0 1 1.2 0.8115775
228 3 0 0 1.5 0.5842653
229 3 0 1 3 0.8115775
230 3 0 1 0.2 0.8115775
231 3 0 1 1.2 0.8115775
232 3 0 0 0.6 0.5842653
233 3 0 0 0 0.5842653
234 3 0 0 0 0.5842653
235 3 1 1 4 0.8505245
236 3 1 1 1.6 0.8505245
237 3 1 1 1.2 0.8505245
238 3 0 1 2.1 0.8115775
239 3 0 1 1.9 0.8115775
240 3 0 1 5.6 0.8115775
241 3 0 1 1.2 0.8115775
242 3 0 1 0.8 0.8115775
243 3 0 1 3.4 0.8115775
244 3 0 1 2 0.8115775
245 3 0 1 2.2 0.8115775
246 3 0 1 1.4 0.8115775
247 3 0 1 2.8 0.8115775
248 3 0 1 0 0.8115775
249 3 0 1 3.2 0.8115775
250 3 0 0 1.2 0.5842653
251 3 1 1 3.1 0.8505245
252 3 0 0 0.4 0.5842653
253 3 0 1 2 0.8115775
254 3 0 0 0 0.5842653
255 3 0 0 0 0.5842653
256 3 0 1 0 0.8115775
257 3 1 0 0.1 0.6499309
258 3 0 1 0 0.8115775
259 3 0 1 1 0.8115775
260 3 0 1 1.2 0.8115775
261 3 0 1 4.2 0.8115775
262 3 0 0 0 0.5842653
263 3 0 1 1.6 0.8115775
264 3 0 1 2.6 0.8115775
265 3 0 1 0.9 0.8115775
266 3 0 0 0 0.5842653
267 3 0 0 0 0.5842653
268 3 0 1 0.5 0.8115775
269 3 0 0 0 0.5842653
270 3 1 1 0 0.8505245
271 3 0 1 1 0.8115775
272 3 0 0 0.1 0.5842653
273 3 0 0 0 0.5842653
274 3 0 1 1.8 0.8115775
275 3 0 1 0.8 0.8115775
276 3 0 0 3 0.5842653
277 3 0 0 0.2 0.5842653
278 3 0 1 0 0.8115775
279 3 0 0 0 0.5842653
280 3 0 1 2.8 0.8115775
281 3 0 1 0 0.8115775
282 3 0 1 0 0.8115775
283 3 0 1 2.5 0.8115775
284 3 1 1 3 0.8505245
285 3 0 0 1.2 0.5842653
286 3 0 0 1.5 0.5842653
287 3 0 0 0 0.5842653
288 3 0 0 0.6 0.5842653
289 3 0 1 1.8 0.8115775
290 3 0 0 0 0.5842653
291 3 0 1 0 0.8115775
292 3 0 1 2 0.8115775
293 3 0 1 0 0.8115775
294 3 0 1 1.2 0.8115775
295 3 0 1 1.6 0.8115775
296 3 0 0 1.4 0.5842653
297 3 0 1 0 0.8115775
> class<-table(data_logistik$Y,data_logistik$yp_hat>0.5)
> class
TRUE
0 96
0.1 6
0.2 12
0.3 3
0.4 8
0.5 5
0.6 14
0.7 1
0.8 13
0.9 3
1 13
1.1 2
1.2 17
1.3 1
1.4 13
1.5 5
1.6 11
1.8 10
1.9 5
2 9
2.1 1
2.2 4
2.3 2
2.4 3
2.5 2
2.6 6
2.8 6
2.9 1
3 5
3.1 1
3.2 2
3.4 3
3.5 1
3.6 4
3.8 1
4 3
4.2 2
4.4 1
5.6 1
6.2 14 HASIL PEMBAHASAN dan KESIMPULAN
Setelah melakukan analisis regresi logistik, diperoleh hasil-hasil berikut:
4.1 Analisa Regresi Logistik
> Berdasarkan hasil output RStudio dapat dibentuk model regresi logistik, yaitu:
+
+ g(x)= 0.02219 + 0.10604X1 + 0.27843X2 + 1.11999X3
+
+ Seluruh nilai koefisien duga pada variabel independen bernilai positif. Sehingga penambahan variabel X1 (sakit jantung), X2 (kecepatan gula darah), dan X3 (kondisi) semua bisa menaikkan terjadinya depresi ST.
Error: <text>:1:13: unexpected symbol
1: Berdasarkan hasil
^4.2 Uji Signifikansi Keseluruhan Model
> Berdasarkan hasil output RStudio diperoleh :
+ G2 = 22.37182728
+ chisquare tabel = 5.99
+
+ nilai G2 > nilai chisquare maka H0 ditolak
+ Sehingga dapat disimpulkan terdapat satu variabel yang berpengaruh terhadap terjadinya depresi ST.
Error: <text>:1:13: unexpected symbol
1: Berdasarkan hasil
^4.3 Asumsi Multikolinieritas
> Berdasarkan hasil output RStudio diperoleh nilai :
+ VIF X1 = 1.164896,
+ VIF X2 = 1.006824,
+ VIF X3 = 1.157597.
+ VIF X1, X2, dan, X3 < 10 maka tidak terjadi multikolinieritas.
Error: <text>:1:13: unexpected symbol
1: Berdasarkan hasil
^4.4 Uji Parsial Parameter Model
> Dapat diketahui variabel X3 (kondisi) memiliki p-value < alpha (0.001) maka H0 ditolak.
+ Sehingga dapat disimpulkan bahwa kondisi berpengaruh signifikan terhadap terjadinya depresi ST.
Error: <text>:1:7: unexpected symbol
1: Dapat diketahui
^4.5 R square
> Didapat nilai R square adalah 0.05985.
+ Dapat disimpulkan bahwa variabel independen (X1, X2, dan X3) dapat menjelaskan 5.985% potensi penyebab depresi ST.
Error: <text>:1:9: unexpected symbol
1: Didapat nilai
^4.6 Odds Ratio
> Variabel X1 (sakit jantung) disimpulkan bahwa jika sakit jantung bertambah 1 maka kecenderungan terjadinya depresi ST meningkat 1.111867 kali lipat.
+
+ Variabel X2 (kecepatan gula darah) disimpulkan bahwa jika kecepatan gula darah bertambah 1 mg/dl maka kecenderungan terjadinya depresi ST meningkat 1.321051 kali lipat.
+
+ Variabel X3 (kondisi) disimpulkan bahwa jika kondisi bertambah 1 disease maka kecenderungan terjadinya depresi ST meningkat 3.064810 kali lipat.
Error: <text>:1:10: unexpected symbol
1: Variabel X1
^5 DAFTAR PUSTAKA
Muflihah, I. Z. (2017). ANALISIS FINANCIAL DISTRESS PERUSAHAAN MANUFAKTUR DI INDONESIA dengan REGRESI LOGISTIK. Majalah Ekonomi, 22(2), 254–269. Retrieved from https://jurnal.unipasby.ac.id/index.php/majalah_ekonomi/article/view/1020
Ghozali, Imam. 2013. Aplikasi Analisis Multivariate dengan Program IBM SPSS 21.Semarang: Badan Penerbit Universitas Diponegoro
Kustituanto,B.,& Badrudin,R. 1994. Statistika 1 (Deskriptif). Jakarta:Gunadarma.