ADKrevisi_Reglog Ordinal

#package

library(ggplot2)
library(tidyverse)

## ── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
## ✔ dplyr     1.1.2     ✔ readr     2.1.4
## ✔ forcats   1.0.0     ✔ stringr   1.5.0
## ✔ lubridate 1.9.2     ✔ tibble    3.2.1
## ✔ purrr     1.0.1     ✔ tidyr     1.3.0
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag()    masks stats::lag()
## ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errors

library(gmodels)
library(ggmosaic)
library(caret)

## Loading required package: lattice
## 
## Attaching package: 'caret'
## 
## The following object is masked from 'package:purrr':
## 
##     lift

library(rpart)
library(rpart.plot)
#library(c50)
library(tidyr)
library(dplyr)
#library(knitr)
library(rsample)
library(readxl)
library(MASS)

## 
## Attaching package: 'MASS'
## 
## The following object is masked from 'package:dplyr':
## 
##     select

#input data

Data_Pendapatan <- read_excel("~/ADK/Data Latihan.xlsx", sheet="Pendapatan")
head(Data_Pendapatan)

## # A tibble: 6 × 7
##   Pendidikan Klasifikasi.Daerah Jenis.Kelamin Status.KRT  Usia Status.Perkawinan
##        <dbl>              <dbl>         <dbl>      <dbl> <dbl>             <dbl>
## 1          0                  0             1          1    49                 1
## 2          0                  0             1          1    47                 1
## 3          0                  0             1          1    61                 1
## 4          1                  0             0          0    23                 0
## 5          0                  0             1          1    27                 1
## 6          0                  0             1          1    50                 1
## # ℹ 1 more variable: Pendapatan <dbl>

#merubah variabel ke faktor

Data_Pendapatan$Pendidikan<- as.factor(Data_Pendapatan$Pendidikan)                         
Data_Pendapatan$Klasifikasi.Daerah<- as.factor(Data_Pendapatan$Klasifikasi.Daerah)
Data_Pendapatan$Jenis.Kelamin<-as.factor(Data_Pendapatan$Jenis.Kelamin)
Data_Pendapatan$Status.KRT<-as.factor(Data_Pendapatan$Status.KRT)
Data_Pendapatan$Status.Perkawinan<-as.factor(Data_Pendapatan$Status.Perkawinan)
Data_Pendapatan$Pendapatan<-as.factor(Data_Pendapatan$Pendapatan)

Data_Pendapatan

## # A tibble: 1,000 × 7
##    Pendidikan Klasifikasi.Daerah Jenis.Kelamin Status.KRT  Usia
##    <fct>      <fct>              <fct>         <fct>      <dbl>
##  1 0          0                  1             1             49
##  2 0          0                  1             1             47
##  3 0          0                  1             1             61
##  4 1          0                  0             0             23
##  5 0          0                  1             1             27
##  6 0          0                  1             1             50
##  7 0          1                  1             1             46
##  8 0          1                  1             1             49
##  9 0          1                  0             0             45
## 10 0          1                  1             0             43
## # ℹ 990 more rows
## # ℹ 2 more variables: Status.Perkawinan <fct>, Pendapatan <fct>

str(Data_Pendapatan)                          

## tibble [1,000 × 7] (S3: tbl_df/tbl/data.frame)
##  $ Pendidikan        : Factor w/ 2 levels "0","1": 1 1 1 2 1 1 1 1 1 1 ...
##  $ Klasifikasi.Daerah: Factor w/ 2 levels "0","1": 1 1 1 1 1 1 2 2 2 2 ...
##  $ Jenis.Kelamin     : Factor w/ 2 levels "0","1": 2 2 2 1 2 2 2 2 1 2 ...
##  $ Status.KRT        : Factor w/ 2 levels "0","1": 2 2 2 1 2 2 2 2 1 1 ...
##  $ Usia              : num [1:1000] 49 47 61 23 27 50 46 49 45 43 ...
##  $ Status.Perkawinan : Factor w/ 2 levels "0","1": 2 2 2 1 2 2 2 2 2 1 ...
##  $ Pendapatan        : Factor w/ 4 levels "1","2","3","4": 2 2 1 2 2 2 4 1 1 1 ...

summary(Data_Pendapatan)                          

##  Pendidikan Klasifikasi.Daerah Jenis.Kelamin Status.KRT      Usia      
##  0:874      0:196              0:309         0:407      Min.   :14.00  
##  1:126      1:804              1:691         1:593      1st Qu.:30.00  
##                                                         Median :39.00  
##                                                         Mean   :39.89  
##                                                         3rd Qu.:49.00  
##                                                         Max.   :80.00  
##  Status.Perkawinan Pendapatan
##  0:262             1:279     
##  1:738             2:568     
##                    3:133     
##                    4: 20     
##                              
## 

library(arsenal)

## 
## Attaching package: 'arsenal'

## The following object is masked from 'package:lubridate':
## 
##     is.Date

tab<-tableby(Pendapatan~ .,data=Data_Pendapatan)
summary(tab,text=TRUE)

## 
## 
## |                   |    1 (N=279)    |    2 (N=568)    |    3 (N=133)    |    4 (N=20)     | Total (N=1000)  | p value|
## |:------------------|:---------------:|:---------------:|:---------------:|:---------------:|:---------------:|-------:|
## |Pendidikan         |                 |                 |                 |                 |                 | < 0.001|
## |-  0               |   268 (96.1%)   |   524 (92.3%)   |   74 (55.6%)    |    8 (40.0%)    |   874 (87.4%)   |        |
## |-  1               |    11 (3.9%)    |    44 (7.7%)    |   59 (44.4%)    |   12 (60.0%)    |   126 (12.6%)   |        |
## |Klasifikasi.Daerah |                 |                 |                 |                 |                 | < 0.001|
## |-  0               |   66 (23.7%)    |   128 (22.5%)   |    2 (1.5%)     |    0 (0.0%)     |   196 (19.6%)   |        |
## |-  1               |   213 (76.3%)   |   440 (77.5%)   |   131 (98.5%)   |   20 (100.0%)   |   804 (80.4%)   |        |
## |Jenis.Kelamin      |                 |                 |                 |                 |                 | < 0.001|
## |-  0               |   132 (47.3%)   |   128 (22.5%)   |   45 (33.8%)    |    4 (20.0%)    |   309 (30.9%)   |        |
## |-  1               |   147 (52.7%)   |   440 (77.5%)   |   88 (66.2%)    |   16 (80.0%)    |   691 (69.1%)   |        |
## |Status.KRT         |                 |                 |                 |                 |                 | < 0.001|
## |-  0               |   141 (50.5%)   |   208 (36.6%)   |   54 (40.6%)    |    4 (20.0%)    |   407 (40.7%)   |        |
## |-  1               |   138 (49.5%)   |   360 (63.4%)   |   79 (59.4%)    |   16 (80.0%)    |   593 (59.3%)   |        |
## |Usia               |                 |                 |                 |                 |                 | < 0.001|
## |-  Mean (SD)       | 42.957 (14.304) | 38.165 (12.304) | 39.910 (10.340) | 46.000 (5.554)  | 39.891 (12.753) |        |
## |-  Range           | 15.000 - 80.000 | 14.000 - 78.000 | 19.000 - 66.000 | 37.000 - 56.000 | 14.000 - 80.000 |        |
## |Status.Perkawinan  |                 |                 |                 |                 |                 |   0.007|
## |-  0               |   91 (32.6%)    |   138 (24.3%)   |   32 (24.1%)    |    1 (5.0%)     |   262 (26.2%)   |        |
## |-  1               |   188 (67.4%)   |   430 (75.7%)   |   101 (75.9%)   |   19 (95.0%)    |   738 (73.8%)   |        |

#pembetukan model

#Partisi data

set.seed(123)

acak<-createDataPartition(Data_Pendapatan$Pendapatan,p=0.8, list=FALSE)
data_train<-Data_Pendapatan[acak, ]
data_test<-Data_Pendapatan[-acak,]

summary(data_train)

##  Pendidikan Klasifikasi.Daerah Jenis.Kelamin Status.KRT      Usia      
##  0:700      0:164              0:243         0:329      Min.   :15.00  
##  1:102      1:638              1:559         1:473      1st Qu.:30.00  
##                                                         Median :39.00  
##                                                         Mean   :39.57  
##                                                         3rd Qu.:48.00  
##                                                         Max.   :80.00  
##  Status.Perkawinan Pendapatan
##  0:219             1:224     
##  1:583             2:455     
##                    3:107     
##                    4: 16     
##                              
## 

summary(data_test)

##  Pendidikan Klasifikasi.Daerah Jenis.Kelamin Status.KRT      Usia      
##  0:174      0: 32              0: 66         0: 78      Min.   :14.00  
##  1: 24      1:166              1:132         1:120      1st Qu.:32.00  
##                                                         Median :40.00  
##                                                         Mean   :41.18  
##                                                         3rd Qu.:49.75  
##                                                         Max.   :78.00  
##  Status.Perkawinan Pendapatan
##  0: 43             1: 55     
##  1:155             2:113     
##                    3: 26     
##                    4:  4     
##                              
## 

#model

model_pendapatan <- polr(Pendapatan~., method="logistic", data=data_train, Hess=T)
summary(model_pendapatan)

## Call:
## polr(formula = Pendapatan ~ ., data = data_train, Hess = T, method = "logistic")
## 
## Coefficients:
##                        Value Std. Error t value
## Pendidikan1          2.52138   0.237066  10.636
## Klasifikasi.Daerah1  0.56414   0.175636   3.212
## Jenis.Kelamin1       0.54822   0.199061   2.754
## Status.KRT1          0.64226   0.212620   3.021
## Usia                -0.03221   0.006865  -4.692
## Status.Perkawinan1   0.53006   0.180398   2.938
## 
## Intercepts:
##     Value   Std. Error t value
## 1|2 -0.5217  0.3040    -1.7161
## 2|3  2.6805  0.3220     8.3249
## 3|4  5.1927  0.4173    12.4437
## 
## Residual Deviance: 1454.495 
## AIC: 1472.495

#uji multicol

library(car)

## Loading required package: carData

## 
## Attaching package: 'car'

## The following object is masked from 'package:dplyr':
## 
##     recode

## The following object is masked from 'package:purrr':
## 
##     some

vif(model_pendapatan)

##         Pendidikan Klasifikasi.Daerah      Jenis.Kelamin         Status.KRT 
##           1.045494           1.026166           1.637565           2.146962 
##               Usia  Status.Perkawinan 
##           1.467146           1.279745

#uji gof

library(generalhoslem)

## Loading required package: reshape

## 
## Attaching package: 'reshape'

## The following object is masked from 'package:lubridate':
## 
##     stamp

## The following object is masked from 'package:dplyr':
## 
##     rename

## The following objects are masked from 'package:tidyr':
## 
##     expand, smiths

lipsitz.test(model_pendapatan)

## 
##  Lipsitz goodness of fit test for ordinal response models
## 
## data:  formula:  Pendapatan ~ Pendidikan + Klasifikasi.Daerah + Jenis.Kelamin + formula:      Status.KRT + Usia + Status.Perkawinan
## LR statistic = 25.252, df = 9, p-value = 0.002705

#partial test

coefmodel <- c(-model_pendapatan$coefficients)
koefisien<-coef(summary(model_pendapatan))

#menghitung p-value
p<-pnorm(abs(koefisien[,"t value"]),lower.tail = FALSE)*2
(ctabel<-cbind <-cbind(round(koefisien,4),"p-value"=round(p,4)))

##                       Value Std. Error t value p-value
## Pendidikan1          2.5214     0.2371 10.6358  0.0000
## Klasifikasi.Daerah1  0.5641     0.1756  3.2120  0.0013
## Jenis.Kelamin1       0.5482     0.1991  2.7540  0.0059
## Status.KRT1          0.6423     0.2126  3.0207  0.0025
## Usia                -0.0322     0.0069 -4.6924  0.0000
## Status.Perkawinan1   0.5301     0.1804  2.9383  0.0033
## 1|2                 -0.5217     0.3040 -1.7161  0.0862
## 2|3                  2.6805     0.3220  8.3249  0.0000
## 3|4                  5.1927     0.4173 12.4437  0.0000

#prediksi pada data test

predict_prob=predict(model_pendapatan, data_test, type="prob")
predict_prob

##              1         2          3            4
## 1   0.33989193 0.5869007 0.06684284 0.0063644817
## 2   0.09094504 0.6200168 0.25712374 0.0319143966
## 3   0.34715626 0.5817922 0.06488758 0.0061639731
## 4   0.22654704 0.6515250 0.11079339 0.0111345959
## 5   0.45852474 0.4956463 0.04194932 0.0038796221
## 6   0.27736764 0.6268207 0.08729227 0.0085193724
## 7   0.45053826 0.5022034 0.04325224 0.0040061223
## 8   0.42708974 0.5211741 0.04733153 0.0044046568
## 9   0.16596916 0.6643290 0.15339845 0.0163033801
## 10  0.47871059 0.4788780 0.03883282 0.0035785796
## 11  0.41888599 0.5277037 0.04885576 0.0045545422
## 12  0.17047630 0.6643126 0.14941636 0.0157947308
## 13  0.19957107 0.6601811 0.12719275 0.0130550895
## 14  0.23224118 0.6492378 0.10773556 0.0107854394
## 15  0.49064332 0.4688438 0.03710068 0.0034121918
## 16  0.16155797 0.6641529 0.15746103 0.0168281297
## 17  0.17978138 0.6637024 0.14169240 0.0148238350
## 18  0.06370555 0.5621674 0.32789533 0.0462316915
## 19  0.45053826 0.5022034 0.04325224 0.0040061223
## 20  0.02126057 0.3268849 0.52003914 0.1318153519
## 21  0.31160469 0.6059504 0.07521149 0.0072334056
## 22  0.27172149 0.6299826 0.08953368 0.0087621991
## 23  0.21281567 0.6564151 0.11871699 0.0120522620
## 24  0.14889167 0.6624712 0.17013321 0.0185039632
## 25  0.22619708 0.6516608 0.11098546 0.0111566195
## 26  0.15302013 0.6632236 0.16582834 0.0179279092
## 27  0.36846966 0.5663638 0.05954561 0.0056209048
## 28  0.26846565 0.6317649 0.09086262 0.0089068110
## 29  0.28130549 0.6245642 0.08577461 0.0083557183
## 30  0.02026991 0.3169006 0.52533696 0.1374925487
## 31  0.16596916 0.6643290 0.15339845 0.0163033801
## 32  0.17978138 0.6637024 0.14169240 0.0148238350
## 33  0.27736764 0.6268207 0.08729227 0.0085193724
## 34  0.17508015 0.6641037 0.14551446 0.0153017039
## 35  0.04556406 0.4944027 0.39541220 0.0646210140
## 36  0.04418351 0.4877719 0.40144897 0.0665956573
## 37  0.03509392 0.4369870 0.44477770 0.0831413986
## 38  0.31855620 0.6014032 0.07303489 0.0070057152
## 39  0.44665532 0.5053745 0.04390095 0.0040692477
## 40  0.27172149 0.6299826 0.08953368 0.0087621991
## 41  0.71705691 0.2671478 0.01449563 0.0012996978
## 42  0.73925761 0.2465998 0.01298064 0.0011619105
## 43  0.20746910 0.6580563 0.12203270 0.0124419135
## 44  0.29756515 0.6148343 0.07987519 0.0077253407
## 45  0.12251660 0.6518963 0.20251080 0.0230763108
## 46  0.30807312 0.6082240 0.07634991 0.0073529734
## 47  0.27095755 0.6304036 0.08984305 0.0087958219
## 48  0.22095233 0.6536287 0.11392405 0.0114949243
## 49  0.13329055 0.6575563 0.18815788 0.0209952352
## 50  0.43464551 0.5151093 0.04597365 0.0042715776
## 51  0.21511971 0.6556604 0.11732969 0.0118902277
## 52  0.45466282 0.4988228 0.04257417 0.0039402413
## 53  0.41461690 0.5310778 0.04967039 0.0046348677
## 54  0.28090167 0.6247975 0.08592857 0.0083722920
## 55  0.42674716 0.5214479 0.04739412 0.0044108014
## 56  0.23803439 0.6467689 0.10474960 0.0104471161
## 57  0.26846565 0.6317649 0.09086262 0.0089068110
## 58  0.28387032 0.6230726 0.08480553 0.0082515393
## 59  0.34641647 0.5823161 0.06508340 0.0061840117
## 60  0.40329027 0.5399469 0.05190667 0.0048561600
## 61  0.27456003 0.6284041 0.08839700 0.0086388872
## 62  0.28090167 0.6247975 0.08592857 0.0083722920
## 63  0.36890388 0.5660432 0.05944247 0.0056104871
## 64  0.23803439 0.6467689 0.10474960 0.0104471161
## 65  0.16596916 0.6643290 0.15339845 0.0163033801
## 66  0.07855333 0.5984520 0.28574846 0.0372462524
## 67  0.27095755 0.6304036 0.08984305 0.0087958219
## 68  0.15724192 0.6637843 0.16160429 0.0173694711
## 69  0.16596916 0.6643290 0.15339845 0.0163033801
## 70  0.16155797 0.6641529 0.15746103 0.0168281297
## 71  0.38026744 0.5575684 0.05681787 0.0053462450
## 72  0.46653259 0.4890267 0.04068364 0.0037571013
## 73  0.02026991 0.3169006 0.52533696 0.1374925487
## 74  0.05474735 0.5327156 0.35866117 0.0538759151
## 75  0.38449844 0.5543738 0.05587599 0.0052518158
## 76  0.46262786 0.4922600 0.04129583 0.0038163179
## 77  0.20222235 0.6595094 0.12542427 0.0128439986
## 78  0.40727088 0.5368441 0.05110803 0.0047769979
## 79  0.25916136 0.6366828 0.09481598 0.0093398483
## 80  0.46662023 0.4889540 0.04067000 0.0037557834
## 81  0.47902691 0.4786131 0.03878589 0.0035740627
## 82  0.43113554 0.5179325 0.04659910 0.0043328228
## 83  0.28465645 0.6226121 0.08451151 0.0082199798
## 84  0.40760895 0.5365799 0.05104085 0.0047703456
## 85  0.42634451 0.5217696 0.04746780 0.0044180360
## 86  0.19707529 0.6607733 0.12889246 0.0132589035
## 87  0.41461690 0.5310778 0.04967039 0.0046348677
## 88  0.07183710 0.5836855 0.30360654 0.0408709027
## 89  0.49064332 0.4688438 0.03710068 0.0034121918
## 90  0.36846966 0.5663638 0.05954561 0.0056209048
## 91  0.24991602 0.6412930 0.09898936 0.0098016624
## 92  0.32873670 0.5945812 0.06999258 0.0066894668
## 93  0.26846565 0.6317649 0.09086262 0.0089068110
## 94  0.15302013 0.6632236 0.16582834 0.0179279092
## 95  0.28786371 0.6207171 0.08332622 0.0080929882
## 96  0.24392621 0.6441200 0.10183454 0.0101192969
## 97  0.24392621 0.6441200 0.10183454 0.0101192969
## 98  0.13705608 0.6590692 0.18353150 0.0203432319
## 99  0.24392621 0.6441200 0.10183454 0.0101192969
## 100 0.37270581 0.5632254 0.05854852 0.0055203013
## 101 0.23224118 0.6492378 0.10773556 0.0107854394
## 102 0.01430860 0.2487227 0.55183273 0.1851359395
## 103 0.10976201 0.6421873 0.22199804 0.0260526150
## 104 0.39137156 0.5491410 0.05438471 0.0051027344
## 105 0.44257719 0.5086927 0.04459336 0.0041367300
## 106 0.42245620 0.5248693 0.04818587 0.0044886033
## 107 0.34342277 0.5844279 0.06588339 0.0062659748
## 108 0.38059748 0.5573200 0.05674373 0.0053388042
## 109 0.17508015 0.6641037 0.14551446 0.0153017039
## 110 0.14889167 0.6624712 0.17013321 0.0185039632
## 111 0.15302013 0.6632236 0.16582834 0.0179279092
## 112 0.16155797 0.6641529 0.15746103 0.0168281297
## 113 0.30124924 0.6125439 0.07861502 0.0075918636
## 114 0.14485554 0.6615275 0.17451883 0.0190981670
## 115 0.23224118 0.6492378 0.10773556 0.0107854394
## 116 0.41461690 0.5310778 0.04967039 0.0046348677
## 117 0.45053826 0.5022034 0.04325224 0.0040061223
## 118 0.55895472 0.4099506 0.02849904 0.0025956108
## 119 0.27483901 0.6282477 0.08828637 0.0086269035
## 120 0.36114393 0.5717369 0.06131877 0.0058004013
## 121 0.02299337 0.3435494 0.51054383 0.1229133549
## 122 0.07183710 0.5836855 0.30360654 0.0408709027
## 123 0.39907098 0.5432185 0.05276876 0.0049417786
## 124 0.06565419 0.5677306 0.32178340 0.0448318633
## 125 0.17978138 0.6637024 0.14169240 0.0148238350
## 126 0.39556304 0.5459246 0.05349803 0.0050143412
## 127 0.23803439 0.6467689 0.10474960 0.0104471161
## 128 0.05898657 0.5474989 0.34353007 0.0499844737
## 129 0.05996905 0.5507002 0.34017377 0.0491570221
## 130 0.01786314 0.2911406 0.53749790 0.1534983909
## 131 0.45451454 0.4989445 0.04259835 0.0039425891
## 132 0.06080023 0.5533463 0.33737671 0.0484767921
## 133 0.18458060 0.6631090 0.13794973 0.0143606721
## 134 0.40329027 0.5399469 0.05190667 0.0048561600
## 135 0.04995616 0.5139051 0.37711901 0.0590196865
## 136 0.62119536 0.3546030 0.02219451 0.0020071335
## 137 0.32474159 0.5972806 0.07116660 0.0068112315
## 138 0.24392621 0.6441200 0.10183454 0.0101192969
## 139 0.41510198 0.5306953 0.04957707 0.0046256581
## 140 0.27095755 0.6304036 0.08984305 0.0087958219
## 141 0.23803439 0.6467689 0.10474960 0.0104471161
## 142 0.19925205 0.6602590 0.12740810 0.0130808615
## 143 0.17047630 0.6643126 0.14941636 0.0157947308
## 144 0.29795524 0.6145932 0.07974048 0.0077110529
## 145 0.22945396 0.6503752 0.10921663 0.0109542133
## 146 0.47064508 0.4856103 0.04004887 0.0036957888
## 147 0.78317279 0.2056922 0.01022275 0.0009122662
## 148 0.53532840 0.4305715 0.03124550 0.0028546055
## 149 0.14485554 0.6615275 0.17451883 0.0190981670
## 150 0.25600312 0.6382900 0.09621303 0.0094939025
## 151 0.34310711 0.5846497 0.06596846 0.0062746996
## 152 0.45852474 0.4956463 0.04194932 0.0038796221
## 153 0.59772329 0.3756330 0.02442894 0.0022147441
## 154 0.10932412 0.6417869 0.22272218 0.0261667667
## 155 0.20476669 0.6588248 0.12376200 0.0126464808
## 156 0.16596916 0.6643290 0.15339845 0.0163033801
## 157 0.25229513 0.6401344 0.09789083 0.0096796313
## 158 0.49064332 0.4688438 0.03710068 0.0034121918
## 159 0.33184489 0.5924620 0.06909632 0.0065967418
## 160 0.31077648 0.6064859 0.07547646 0.0072612050
## 161 0.47067814 0.4855828 0.04004381 0.0036953003
## 162 0.01386131 0.2429735 0.55312021 0.1900449810
## 163 0.62874533 0.3478022 0.02150882 0.0019436317
## 164 0.40329027 0.5399469 0.05190667 0.0048561600
## 165 0.55895472 0.4099506 0.02849904 0.0025956108
## 166 0.01730668 0.2848615 0.54010086 0.1577309180
## 167 0.32873670 0.5945812 0.06999258 0.0066894668
## 168 0.42674716 0.5214479 0.04739412 0.0044108014
## 169 0.20444149 0.6589145 0.12397259 0.0126714564
## 170 0.27095755 0.6304036 0.08984305 0.0087958219
## 171 0.21545729 0.6555475 0.11712846 0.0118667733
## 172 0.30124924 0.6125439 0.07861502 0.0075918636
## 173 0.35777089 0.5741868 0.06215679 0.0058854969
## 174 0.22095233 0.6536287 0.11392405 0.0114949243
## 175 0.23224118 0.6492378 0.10773556 0.0107854394
## 176 0.32474159 0.5972806 0.07116660 0.0068112315
## 177 0.22095233 0.6536287 0.11392405 0.0114949243
## 178 0.22654704 0.6515250 0.11079339 0.0111345959
## 179 0.17047630 0.6643126 0.14941636 0.0157947308
## 180 0.21545729 0.6555475 0.11712846 0.0118667733
## 181 0.24392621 0.6441200 0.10183454 0.0101192969
## 182 0.28130549 0.6245642 0.08577461 0.0083557183
## 183 0.17047630 0.6643126 0.14941636 0.0157947308
## 184 0.37613718 0.5606669 0.05775550 0.0054404569
## 185 0.23803439 0.6467689 0.10474960 0.0104471161
## 186 0.30137514 0.6124651 0.07857243 0.0075873593
## 187 0.12961293 0.6558557 0.19286365 0.0216676732
## 188 0.20476669 0.6588248 0.12376200 0.0126464808
## 189 0.26464136 0.6338185 0.09245901 0.0090811599
## 190 0.03558130 0.4400642 0.44229651 0.0820579892
## 191 0.19447504 0.6613477 0.13070051 0.0134767203
## 192 0.23224118 0.6492378 0.10773556 0.0107854394
## 193 0.17978138 0.6637024 0.14169240 0.0148238350
## 194 0.15724192 0.6637843 0.16160429 0.0173694711
## 195 0.37303323 0.5629819 0.05847229 0.0055126196
## 196 0.20476669 0.6588248 0.12376200 0.0126464808
## 197 0.40362728 0.5396848 0.05183851 0.0048493980
## 198 0.30807312 0.6082240 0.07634991 0.0073529734

head(data.frame(predict_prob, data_test$Pendapatan), 20)

##            X1        X2         X3          X4 data_test.Pendapatan
## 1  0.33989193 0.5869007 0.06684284 0.006364482                    2
## 2  0.09094504 0.6200168 0.25712374 0.031914397                    2
## 3  0.34715626 0.5817922 0.06488758 0.006163973                    2
## 4  0.22654704 0.6515250 0.11079339 0.011134596                    1
## 5  0.45852474 0.4956463 0.04194932 0.003879622                    1
## 6  0.27736764 0.6268207 0.08729227 0.008519372                    3
## 7  0.45053826 0.5022034 0.04325224 0.004006122                    2
## 8  0.42708974 0.5211741 0.04733153 0.004404657                    2
## 9  0.16596916 0.6643290 0.15339845 0.016303380                    1
## 10 0.47871059 0.4788780 0.03883282 0.003578580                    1
## 11 0.41888599 0.5277037 0.04885576 0.004554542                    3
## 12 0.17047630 0.6643126 0.14941636 0.015794731                    2
## 13 0.19957107 0.6601811 0.12719275 0.013055090                    3
## 14 0.23224118 0.6492378 0.10773556 0.010785439                    1
## 15 0.49064332 0.4688438 0.03710068 0.003412192                    1
## 16 0.16155797 0.6641529 0.15746103 0.016828130                    2
## 17 0.17978138 0.6637024 0.14169240 0.014823835                    2
## 18 0.06370555 0.5621674 0.32789533 0.046231692                    3
## 19 0.45053826 0.5022034 0.04325224 0.004006122                    2
## 20 0.02126057 0.3268849 0.52003914 0.131815352                    2

#confussion matrix nilai akurasi model dalam melakukan prediksi 0.6111, artinya ketepatan model dalam melakukan prediksi sebesar 61.11%

prediksi.test <- predict(model_pendapatan, data_test, type="class")

data_test$Pendapatan<-as.factor(data_test$Pendapatan)

confusionMatrix(as.factor(prediksi.test), data_test$Pendapatan, positive="1")

## Confusion Matrix and Statistics
## 
##           Reference
## Prediction   1   2   3   4
##          1  11   2   0   0
##          2  43 107  23   2
##          3   1   4   3   2
##          4   0   0   0   0
## 
## Overall Statistics
##                                           
##                Accuracy : 0.6111          
##                  95% CI : (0.5394, 0.6794)
##     No Information Rate : 0.5707          
##     P-Value [Acc > NIR] : 0.1406          
##                                           
##                   Kappa : 0.1738          
##                                           
##  Mcnemar's Test P-Value : NA              
## 
## Statistics by Class:
## 
##                      Class: 1 Class: 2 Class: 3 Class: 4
## Sensitivity           0.20000   0.9469  0.11538   0.0000
## Specificity           0.98601   0.2000  0.95930   1.0000
## Pos Pred Value        0.84615   0.6114  0.30000      NaN
## Neg Pred Value        0.76216   0.7391  0.87766   0.9798
## Prevalence            0.27778   0.5707  0.13131   0.0202
## Detection Rate        0.05556   0.5404  0.01515   0.0000
## Detection Prevalence  0.06566   0.8838  0.05051   0.0000
## Balanced Accuracy     0.59301   0.5735  0.53734   0.5000

#odds ratio untuk interpretasi model, dilakukan dengan menghitung nilai odds ratio

data.frame(coef(model_pendapatan), exp(coef(model_pendapatan)))

##                     coef.model_pendapatan. exp.coef.model_pendapatan..
## Pendidikan1                     2.52137511                  12.4456991
## Klasifikasi.Daerah1             0.56413644                   1.7579291
## Jenis.Kelamin1                  0.54821922                   1.7301692
## Status.KRT1                     0.64226197                   1.9007755
## Usia                           -0.03221301                   0.9683003
## Status.Perkawinan1              0.53005786                   1.6990306

kesimpulan: Tingkat pendidikan, 12.44, Responden pendidikan tinggi memiliki peluang 12.44 kali lebih besar untuk mendapatkan pendapatan yang lebih tinggi dibandingkan lulusan SMA ke bawah.

Jenis kelamin, 1.73 Responden dengan jenis kelamin laki-laki memiliki peluang 1.73 kali lebih besar dibanding responden perempuan untuk mendapatkan pendapatan yang lebih besar.

Status perkawinan, 1.69 Responden dengan status yang sudah menikah memiliki peluang 1.69 kai lebih besar dibanding responden yang belum menikah untuk mendapatkan pendapatan yang lebih besar.