Metodologi Penelitian

Ujian Akhir Semester

1 Latar Belakang

Tujuan dari project ini adalah untuk memprediksi penjualan Busana Pria di Belanda. Dataset terdiri dari Penjualan Toko Busana Pria di Belanda dengan 400 observasi.

2 Persiapan Data

2.1 Packages

Berikut beberapa packages yang digunakan, silakan untuk melakukan instalasi packages terlebih dahulu jika diperlukan.

library(tidyverse)
library(dplyr)
library(ggplot2)
library(ggcorrplot)
library(GGally)
library(lubridate)
library(gridExtra)
library(grid)
library(MLmetrics)
library(scales)
library(nortest)

2.2 Import Data

Data yang digunakan dalam project ini adalah Clothing. Dataset ini terdiri dari 400 data dengan 13 variabel.

Clothing <- read.csv("Clothing.csv")
data.frame("Total.Data" = dim(Clothing)[1],
           "Total.Variabel" = dim(Clothing)[2])

##   Total.Data Total.Variabel
## 1        400             13

names(Clothing)

##  [1] "tsale"   "sales"   "margin"  "nown"    "nfull"   "npart"   "naux"   
##  [8] "hoursw"  "hourspw" "inv1"    "inv2"    "ssize"   "start"

Berikut dilampirkan 10 data teratas.

head(Clothing, 10)

##      tsale     sales margin   nown  nfull  npart   naux hoursw  hourspw
## 1   750000  4411.765     41 1.0000 1.0000 1.0000 1.5357     76 16.75596
## 2  1926395  4280.878     39 2.0000 2.0000 3.0000 1.5357    192 22.49376
## 3  1250000  4166.667     40 1.0000 2.0000 2.2222 1.4091    114 17.19120
## 4   694227  2670.104     40 1.0000 1.0000 1.2833 1.3673    100 21.50260
## 5   750000 15000.000     44 2.0000 1.9556 1.2833 1.3673    104 15.74279
## 6   400000  4444.444     41 2.0000 1.9556 1.2833 1.3673     72 10.89885
## 7  1300000  3250.000     39 1.2228 1.0000 3.0000 4.0000    161 17.45674
## 8   495340  4953.400     28 2.0000 1.9556 1.2833 1.3673     80 12.10984
## 9  1200000  2666.667     41 1.0000 3.0000 2.2222 1.4091    158 20.70420
## 10  495340  6604.533     37 1.0000 1.9556 1.2833 1.0000     87 16.60654
##         inv1     inv2 ssize start
## 1   17166.67 27177.04   170  1984
## 2   17166.67 27177.04   450  1972
## 3  292857.20 71570.55   300  1952
## 4   22207.04 15000.00   260  1966
## 5   22207.04 10000.00    50  1996
## 6   22207.04 22859.85    90  1947
## 7   22207.04 22859.85   400  1993
## 8   22207.04 22859.85   100  1952
## 9  292857.20  5000.00   450  1954
## 10  22207.04 22859.85    75  1973

2.3 Deskripisi Variabel

Berikut beberapa deskripsi dari variabel dataset ini:

Variabel	Deskripsi
tsales	Annual sales in Dutch guilders
sales	Sales per square meter
margin	Gross-profit-margin
nown	Number of owners (managers)
nfull	Number of full-timers
npart	Number of part-timers
naux	Number of helpers (temporary workers)
hoursw	Total number of hours worked
hourspw	Number of hours worked per worker
inv1	Investment in shop-premises
inv2	Investment in automation
ssize	Sales floorspace of the store
start	Year start of business

2.4 Pembersihan Data

2.4.1 Periksa Struktur Data

Berikut adalah struktur dari dataset tersebut.

str(Clothing)

## 'data.frame':    400 obs. of  13 variables:
##  $ tsale  : int  750000 1926395 1250000 694227 750000 400000 1300000 495340 1200000 495340 ...
##  $ sales  : num  4412 4281 4167 2670 15000 ...
##  $ margin : num  41 39 40 40 44 41 39 28 41 37 ...
##  $ nown   : num  1 2 1 1 2 ...
##  $ nfull  : num  1 2 2 1 1.96 ...
##  $ npart  : num  1 3 2.22 1.28 1.28 ...
##  $ naux   : num  1.54 1.54 1.41 1.37 1.37 ...
##  $ hoursw : int  76 192 114 100 104 72 161 80 158 87 ...
##  $ hourspw: num  16.8 22.5 17.2 21.5 15.7 ...
##  $ inv1   : num  17167 17167 292857 22207 22207 ...
##  $ inv2   : num  27177 27177 71571 15000 10000 ...
##  $ ssize  : int  170 450 300 260 50 90 400 100 450 75 ...
##  $ start  : int  1984 1972 1952 1966 1996 1947 1993 1952 1954 1973 ...

rmarkdown::paged_table(Clothing)

2.5 Periksa Data Hilang

Dalam setiap proses pengolahan data kita dapat melakukan pemeriksaan data yang hilang (missing values) sebelum melakukan proses analisa dengan cara sebagai berikut:

colSums(is.na(Clothing))

##   tsale   sales  margin    nown   nfull   npart    naux  hoursw hourspw    inv1 
##       0       0       0       0       0       0       0       0       0       0 
##    inv2   ssize   start 
##       0       0       0

Hasilnya tidak terdapat missing value pada dataset ini.

2.6 Data Duplikat

data.frame("jumlah.seluruh.data"=nrow(Clothing),
           "jumlah.data.unik" = nrow(distinct(Clothing))
           )

##   jumlah.seluruh.data jumlah.data.unik
## 1                 400              400

Hasilnya tidak terdapat data duplikat pada dataset ini.

2.7 Ringkasan Data

Berikut ringkasan data yang dapat dilihat.

summary(Clothing)  

##      tsale             sales           margin           nown       
##  Min.   :  50000   Min.   :  300   Min.   :16.00   Min.   : 1.000  
##  1st Qu.: 495340   1st Qu.: 3904   1st Qu.:37.00   1st Qu.: 1.000  
##  Median : 694227   Median : 5279   Median :39.00   Median : 1.000  
##  Mean   : 833584   Mean   : 6335   Mean   :38.77   Mean   : 1.284  
##  3rd Qu.: 976817   3rd Qu.: 7740   3rd Qu.:41.00   3rd Qu.: 1.295  
##  Max.   :5000000   Max.   :27000   Max.   :66.00   Max.   :10.000  
##      nfull           npart            naux           hoursw     
##  Min.   :1.000   Min.   :1.000   Min.   :1.000   Min.   : 32.0  
##  1st Qu.:1.923   1st Qu.:1.283   1st Qu.:1.333   1st Qu.: 80.0  
##  Median :1.956   Median :1.283   Median :1.367   Median :104.0  
##  Mean   :2.069   Mean   :1.566   Mean   :1.390   Mean   :121.1  
##  3rd Qu.:2.066   3rd Qu.:2.000   3rd Qu.:1.367   3rd Qu.:145.2  
##  Max.   :8.000   Max.   :9.000   Max.   :4.000   Max.   :582.0  
##     hourspw            inv1              inv2            ssize       
##  Min.   : 5.708   Min.   :   1000   Min.   :   350   Min.   :  16.0  
##  1st Qu.:13.541   1st Qu.:  20000   1st Qu.: 10000   1st Qu.:  80.0  
##  Median :17.745   Median :  22207   Median : 22860   Median : 120.0  
##  Mean   :18.955   Mean   :  58257   Mean   : 27829   Mean   : 151.1  
##  3rd Qu.:24.303   3rd Qu.:  62269   3rd Qu.: 22860   3rd Qu.: 190.0  
##  Max.   :43.326   Max.   :1500000   Max.   :400000   Max.   :1214.0  
##      start     
##  Min.   :1945  
##  1st Qu.:1959  
##  Median :1978  
##  Mean   :1978  
##  3rd Qu.:1996  
##  Max.   :2015

3 Analisis Eksplorasi Data

3.1 Matriks Korelasi

ggcorr(Clothing,label = T, size=3, label_size = 3, hjust=0.95)+
  labs(
    title="Matriks Korelasi Data set"
  )+
  theme_minimal()+
  theme(
    plot.title = element_text(hjust = 0.5),
    axis.title=element_text(size=8,face="bold"), 
    axis.text.y=element_blank()
  )

Berdasarkan matriks korelasi diatas, setiap variabel mempunyai pengaruh terhadap tsale. Lalu variabel yang memiliki korelasi paling tinggi dengan tsale adalah hoursw.

3.2 Pencilan Harga Penjualan

ggplot(Clothing, aes(x=tsale)) + 
 geom_histogram(aes(x=tsale, y=..density..), colour="black", fill="white")+
 geom_density(alpha=.2, fill="blue")+
  labs(title="Distribusi Penjualan Tanpa Outlier")+
  theme_minimal()+
  theme(
    plot.title = element_text(hjust = 0.5),
    axis.title=element_text(size=9,face="bold")
  )

Jika dilihat terdapat cukup banyak data yang memiliki tsale yang jauh lebih tinggi dari data lainnya sehingga menyebabkan distribusi penyebaran total volume tidak normal. Sehingga saya memutuskan untuk membuang data dengan tsale yang lebih dari 4 juta karena jika dilihat data yang memiliki tsale yang lebih dari 4 juta sangat sedikit.

Clothing_clean <- Clothing %>% 
  filter(tsale < 4000000)

ggplot(Clothing_clean, aes(x=tsale)) + 
 geom_histogram(aes(y=..density..), colour="black", fill="white",binwidth = 80000)+
 geom_density(alpha=.2, fill="blue")+
   scale_x_continuous(name = "tsale")+
  labs(title="Distribusi Harga Penjualan < 4 Juta")+
  theme_minimal()+
  theme(
    plot.title = element_text(hjust = 0.5),
    axis.title=element_text(size=9,face="bold"), 
    axis.text.y=element_text(margin = margin(l=5)),
    axis.text.x.bottom = element_text(margin = margin(b=5))
  )

4 Pemisahan Data

Sebelum membangun model regresi linear, saya membagi data train:test dengan proporsi 80:20. Proporsi ini yang sudah sering digunakan dalam proses pemodelan machine learning.

4.1 Train vs Test

set.seed(999)
idx_Clothing <- sample(x=nrow(Clothing_clean),size=nrow(Clothing)*0.8)
Clothing_train <- Clothing_clean[idx_Clothing,]
Clothing_test <- Clothing_clean[-idx_Clothing,]

5 Pemodelan : Regresi Linear

5.1 Prediktor Tunggal

Berdasarkan matriks korelasi sebelumnya, variabel hoursw memiliki korelasi paling kuat dengan variabel tsale.

lm_Clothing <- lm(tsale~hoursw,Clothing_train)
summary(lm_Clothing)

## 
## Call:
## lm(formula = tsale ~ hoursw, data = Clothing_train)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -1844749  -216720   -53231   151918  2036572 
## 
## Coefficients:
##             Estimate Std. Error t value Pr(>|t|)    
## (Intercept)  52567.4    42404.9    1.24    0.216    
## hoursw        6389.3      306.4   20.85   <2e-16 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 359300 on 318 degrees of freedom
## Multiple R-squared:  0.5775, Adjusted R-squared:  0.5762 
## F-statistic: 434.7 on 1 and 318 DF,  p-value: < 2.2e-16

Berdasarkan single predictor, model ini hanya memiliki nilai Adjusted R-squared: 0.5762 yang berarti hanya berhasil menangkap variansi dari target sebesar 58%.

Jadi, Model Linear Regression menggunakan single predictor tidak cocok diterapkan untuk menentukan harga penjualan busana pria pada dataset ini.

5.2 Prediktor Ganda

lm_Clothing_all <- lm(tsale~., data=Clothing_train)
summary(lm_Clothing_all)

## 
## Call:
## lm(formula = tsale ~ ., data = Clothing_train)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -1647613  -103129    -1013    92850  1484180 
## 
## Coefficients:
##               Estimate Std. Error t value Pr(>|t|)    
## (Intercept)  7.849e+05  1.370e+06   0.573   0.5671    
## sales        7.591e+01  4.930e+00  15.398   <2e-16 ***
## margin      -2.266e+02  2.853e+03  -0.079   0.9367    
## nown        -4.690e+04  4.165e+04  -1.126   0.2610    
## nfull        9.169e+04  3.652e+04   2.511   0.0126 *  
## npart        4.249e+04  3.720e+04   1.142   0.2542    
## naux         9.222e+04  4.550e+04   2.027   0.0436 *  
## hoursw       5.068e+02  1.285e+03   0.394   0.6936    
## hourspw      1.801e+04  8.267e+03   2.178   0.0301 *  
## inv1        -1.227e-01  1.493e-01  -0.821   0.4120    
## inv2         2.292e-01  3.623e-01   0.633   0.5275    
## ssize        2.265e+03  2.003e+02  11.306   <2e-16 ***
## start       -7.559e+02  6.901e+02  -1.095   0.2742    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 251000 on 307 degrees of freedom
## Multiple R-squared:  0.8009, Adjusted R-squared:  0.7932 
## F-statistic: 102.9 on 12 and 307 DF,  p-value: < 2.2e-16

terlihat nilai Adjusted R-squared:0.793 , hal ini menandakan bahwa model lm_Clothing_all dapat memprediksi 79.3%.

5.3 Stepwise Method

step_backward <- step(lm_Clothing_all,direction = "backward")

## Start:  AIC=7970.04
## tsale ~ sales + margin + nown + nfull + npart + naux + hoursw + 
##     hourspw + inv1 + inv2 + ssize + start
## 
##           Df  Sum of Sq        RSS    AIC
## - margin   1 3.9767e+08 1.9345e+13 7968.0
## - hoursw   1 9.8003e+09 1.9355e+13 7968.2
## - inv2     1 2.5209e+10 1.9370e+13 7968.5
## - inv1     1 4.2522e+10 1.9388e+13 7968.7
## - start    1 7.5619e+10 1.9421e+13 7969.3
## - nown     1 7.9903e+10 1.9425e+13 7969.4
## - npart    1 8.2245e+10 1.9427e+13 7969.4
## <none>                  1.9345e+13 7970.0
## - naux     1 2.5883e+11 1.9604e+13 7972.3
## - hourspw  1 2.9902e+11 1.9644e+13 7973.0
## - nfull    1 3.9722e+11 1.9742e+13 7974.5
## - ssize    1 8.0541e+12 2.7399e+13 8079.4
## - sales    1 1.4940e+13 3.4285e+13 8151.2
## 
## Step:  AIC=7968.05
## tsale ~ sales + nown + nfull + npart + naux + hoursw + hourspw + 
##     inv1 + inv2 + ssize + start
## 
##           Df  Sum of Sq        RSS    AIC
## - hoursw   1 9.7477e+09 1.9355e+13 7966.2
## - inv2     1 2.5230e+10 1.9371e+13 7966.5
## - inv1     1 4.2180e+10 1.9388e+13 7966.7
## - start    1 7.5607e+10 1.9421e+13 7967.3
## - nown     1 7.9735e+10 1.9425e+13 7967.4
## - npart    1 8.1884e+10 1.9427e+13 7967.4
## <none>                  1.9345e+13 7968.0
## - naux     1 2.6467e+11 1.9610e+13 7970.4
## - hourspw  1 2.9862e+11 1.9644e+13 7971.0
## - nfull    1 3.9755e+11 1.9743e+13 7972.6
## - ssize    1 8.0825e+12 2.7428e+13 8077.8
## - sales    1 1.5063e+13 3.4409e+13 8150.3
## 
## Step:  AIC=7966.21
## tsale ~ sales + nown + nfull + npart + naux + hourspw + inv1 + 
##     inv2 + ssize + start
## 
##           Df  Sum of Sq        RSS    AIC
## - inv2     1 2.5949e+10 1.9381e+13 7964.6
## - inv1     1 4.4664e+10 1.9400e+13 7964.9
## - start    1 7.9809e+10 1.9435e+13 7965.5
## <none>                  1.9355e+13 7966.2
## - nown     1 1.4669e+11 1.9502e+13 7966.6
## - npart    1 2.3619e+11 1.9591e+13 7968.1
## - naux     1 5.2588e+11 1.9881e+13 7972.8
## - nfull    1 2.3855e+12 2.1741e+13 8001.4
## - hourspw  1 4.6509e+12 2.4006e+13 8033.1
## - ssize    1 8.1207e+12 2.7476e+13 8076.3
## - sales    1 1.5068e+13 3.4424e+13 8148.5
## 
## Step:  AIC=7964.64
## tsale ~ sales + nown + nfull + npart + naux + hourspw + inv1 + 
##     ssize + start
## 
##           Df  Sum of Sq        RSS    AIC
## - inv1     1 2.5289e+10 1.9406e+13 7963.1
## - start    1 7.4609e+10 1.9456e+13 7963.9
## <none>                  1.9381e+13 7964.6
## - nown     1 1.3428e+11 1.9515e+13 7964.8
## - npart    1 2.2004e+11 1.9601e+13 7966.3
## - naux     1 5.2495e+11 1.9906e+13 7971.2
## - nfull    1 2.4031e+12 2.1784e+13 8000.0
## - hourspw  1 4.7241e+12 2.4105e+13 8032.4
## - ssize    1 8.2971e+12 2.7678e+13 8076.7
## - sales    1 1.5045e+13 3.4426e+13 8146.5
## 
## Step:  AIC=7963.06
## tsale ~ sales + nown + nfull + npart + naux + hourspw + ssize + 
##     start
## 
##           Df  Sum of Sq        RSS    AIC
## - start    1 6.7331e+10 1.9474e+13 7962.2
## <none>                  1.9406e+13 7963.1
## - nown     1 1.4030e+11 1.9547e+13 7963.4
## - npart    1 2.0605e+11 1.9612e+13 7964.4
## - naux     1 5.5560e+11 1.9962e+13 7970.1
## - nfull    1 2.3803e+12 2.1787e+13 7998.1
## - hourspw  1 4.7005e+12 2.4107e+13 8030.5
## - ssize    1 8.2800e+12 2.7686e+13 8074.8
## - sales    1 1.5091e+13 3.4497e+13 8145.1
## 
## Step:  AIC=7962.16
## tsale ~ sales + nown + nfull + npart + naux + hourspw + ssize
## 
##           Df  Sum of Sq        RSS    AIC
## <none>                  1.9474e+13 7962.2
## - nown     1 1.3095e+11 1.9605e+13 7962.3
## - npart    1 1.8971e+11 1.9663e+13 7963.3
## - naux     1 5.4986e+11 2.0024e+13 7969.1
## - nfull    1 2.4645e+12 2.1938e+13 7998.3
## - hourspw  1 4.7089e+12 2.4183e+13 8029.5
## - ssize    1 8.2229e+12 2.7697e+13 8072.9
## - sales    1 1.5100e+13 3.4574e+13 8143.9

summary(step_backward)

## 
## Call:
## lm(formula = tsale ~ sales + nown + nfull + npart + naux + hourspw + 
##     ssize, data = Clothing_train)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -1634385   -99151    -3269    97041  1491549 
## 
## Coefficients:
##               Estimate Std. Error t value Pr(>|t|)    
## (Intercept) -787834.86   76363.79 -10.317  < 2e-16 ***
## sales            75.90       4.88  15.554  < 2e-16 ***
## nown         -30712.84   21204.06  -1.448  0.14850    
## nfull        104779.18   16674.71   6.284 1.11e-09 ***
## npart         45967.90   26366.81   1.743  0.08225 .  
## naux         105213.32   35447.96   2.968  0.00323 ** 
## hourspw       21131.21    2432.83   8.686  < 2e-16 ***
## ssize          2257.65     196.69  11.478  < 2e-16 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 249800 on 312 degrees of freedom
## Multiple R-squared:  0.7996, Adjusted R-squared:  0.7951 
## F-statistic: 177.9 on 7 and 312 DF,  p-value: < 2.2e-16

6 Evaluasi Model

Berdasarkan hasil stepwise backward, formula untuk Multiple Linear Regression yang disarankan yaitu :

lm_Clothing_model <- lm(formula = tsale ~ sales + nown + nfull + npart + naux + hourspw + 
    ssize, data = Clothing_train)

6.1 Linearitas

linearity <- data.frame(residual = lm_Clothing_model$residuals, fitted = Clothing_train$tsale)
linearity %>% 
  ggplot(aes(fitted, residual)) + 
  geom_point() + 
  geom_smooth(method = lm) + 
  geom_hline(aes(yintercept = 0)) + 
     theme_minimal()+
  theme(
    plot.title = element_text(hjust = 0.5),
    axis.title=element_text(size=9,face="bold"), 
    axis.text.y=element_text(margin = margin(l=5)),
    axis.text.x.bottom = element_text(margin = margin(b=5)))

Berdasarkan pengujian tersebut, model yang dibuat belum berhasil menangkap variansi data dengan baik karena masih terlihat error atau jarak data dari mean cukup jauh.

6.2 Normalitas Residual

hist(lm_Clothing_model$residuals)

#shapiro test
shapiro.test(lm_Clothing_model$residuals[0:5000])

## 
##  Shapiro-Wilk normality test
## 
## data:  lm_Clothing_model$residuals[0:5000]
## W = 0.86235, p-value = 2.986e-16

# Anderson-Darling normality test
nortest::ad.test(lm_Clothing_model$residuals)

## 
##  Anderson-Darling normality test
## 
## data:  lm_Clothing_model$residuals
## A = 8.5716, p-value < 2.2e-16

Hipotesis:

• H0: Residual berdistribusi normal
• H1: Residual tidak berdistribusi normal

Jadi, Karena nilai p-value < 0.05 maka keputusannya adalah terima H1, dengan kesimpulan residual dari model tidak berdistribusi normal.

6.3 Heteroskedastisitas

lmtest::bptest(lm_Clothing_model)

## 
##  studentized Breusch-Pagan test
## 
## data:  lm_Clothing_model
## BP = 140.31, df = 7, p-value < 2.2e-16

Hipotesis:

• H0: Data residual Homogen (tidak membentuk sebuah pola)
• H1: Data residual heteros (membentuk sebuah pola)

Jadi, Karena nilai p-value < 0.05 maka keputusannya adalah terima H1, dengan kesimpulan residual dari model bersifat heteros, dapat dilihat membentuk sebuah pola.

7 Prediksi

predict_price <-  predict(
  lm_Clothing_model,
  Clothing_test,
  interval = "confidence",
  level = 0.95
)

Clothing_predict <- cbind(Clothing_test,predict_price)

Clothing_predict %>% 
  select(
  tsale, predict.fit=fit, predict.lower=lwr, predict.upper=upr,
  sales, margin, nown, nfull, npart, naux, hoursw, hourspw, inv1,
  inv2, ssize)

##       tsale predict.fit predict.lower predict.upper     sales margin   nown
## 1    750000   566511.13     509927.43      623094.8  4411.765     41 1.0000
## 3   1250000   998241.05     938778.42     1057703.7  4166.667     40 1.0000
## 12   231000   263618.08     206543.47      320692.7  5775.000     30 1.0000
## 15   330000   662292.54     603891.69      720693.4  8250.000     34 1.0000
## 20   330000   350067.46     306919.45      393215.5  4714.286     41 1.0000
## 25   471000   581879.69     523037.56      640721.8  4957.895     39 1.0000
## 30   694227   790844.75     700236.18      881453.3  5785.225     44 1.0000
## 32   495340   446500.52     393553.66      499447.4  4953.400     41 2.0000
## 39   876000   673050.73     609977.62      736123.8  6257.143     40 1.0000
## 59   549000   399529.22     336477.74      462580.7  4575.000     41 2.0000
## 63   976817  1157535.44    1061895.96     1253174.9 15027.950     43 1.0000
## 64   450000   611885.23     565559.62      658210.8  5000.000     39 1.0000
## 75   301133   483346.77     423212.11      543481.4  8603.800     37 1.0000
## 76   156168   341222.56     282670.94      399774.2  1382.018     37 1.0000
## 78   301133   215570.06     164135.32      267004.8  3011.330     35 1.0000
## 83  1343500  1768292.22    1661012.04     1875572.4 19192.860     42 1.2228
## 84  1290500  1676427.42    1570167.44     1782687.4 18435.710     42 1.2228
## 94   777362   699383.02     643974.35      754791.7  5182.414     32 1.0000
## 95   976817   791963.22     751220.08      832706.4  4884.085     39 1.0000
## 103  495340   472597.92     419507.47      525688.4  4503.091     31 1.0000
## 105  876845   864103.36     801282.84      926923.9  5480.281     42 1.0000
## 107  694227   631676.57     579872.39      683480.8  5833.840     41 2.0000
## 110  694227   642700.13     591986.09      693414.2  4338.919     39 2.0000
## 116  156168    38603.63     -24139.97      101347.2  1952.100     25 1.0000
## 118  976817   783417.62     719462.35      847372.9  9768.170     44 1.0000
## 127  301133   369388.82     310549.59      428228.0  2007.553     28 2.0000
## 134  495340   418265.18     353908.59      482621.8  4953.400     39 2.0000
## 136  976817  1099072.68    1005138.46     1193006.9 13954.530     41 1.0000
## 140  976817   789714.22     748338.52      831089.9  5280.092     41 1.0000
## 141 1378586  1214669.14    1151839.17     1277499.1  7877.634     44 1.0000
## 151  390000   498955.51     456311.92      541599.1  5571.429     42 1.0000
## 152  535000   482557.84     438656.87      526458.8  4458.333     34 1.0000
## 154 1926395  1936548.98    1838243.72     2034854.2 17512.680     41 2.0000
## 187  976817  1685055.27    1564739.14     1805371.4  2382.480     66 1.0000
## 189  694227  1151951.19    1055904.70     1247997.7  9917.528     39 1.0000
## 200  694227   597192.14     508909.94      685474.3  4958.764     44 1.0000
## 201  694227   564645.38     508588.00      620702.8  4958.764     44 1.0000
## 203  301133   394759.47     350304.54      439214.4  3764.163     41 1.0000
## 219 1475000  1280945.06    1220201.68     1341688.4  4402.985     41 1.0000
## 220  795655  1219494.96    1121736.74     1317253.2  1768.122     41 2.0000
## 221  570000   484691.60     437106.89      532276.3  5700.000     40 1.0000
## 230  770200   982477.29     899575.00     1065379.6  2026.842     41 2.0000
## 237  495340   523057.96     481325.79      564790.1  2896.725     40 1.0000
## 245  495340   606476.68     562384.68      650568.7  7620.615     41 1.0000
## 246 1400000  1040360.15     959534.85     1121185.5  7777.778     41 2.0000
## 251  971500   925507.25     847969.18     1003045.3  3736.539     41 2.0000
## 260  976817  1094600.96    1004091.68     1185110.2  2790.906     44 2.0000
## 262  301133   302903.96     257693.17      348114.8  4182.403     31 1.0000
## 270  301133   238725.02     186441.94      291008.1  3764.163     39 1.0000
## 271  750000   857358.82     770717.79      943999.9  3750.000     36 2.0000
## 272  301133   371463.26     326598.63      416327.9  3011.330     39 1.0000
## 274  301133   389601.33     340467.89      438734.8  2737.573     35 1.0000
## 278  976817   985752.63     918060.78     1053444.5 12210.210     39 2.0000
## 283  800000   939434.26     850168.86     1028699.6  5333.333     37 2.0000
## 289  976817   702568.64     628540.89      776596.4  3907.268     41 1.0000
## 290 1076500  1069649.19     983471.73     1155826.6  5053.991     39 1.0000
## 305  694227   494188.19     430892.04      557484.3  6942.270     35 2.0000
## 308  984717  1035699.40     983551.33     1087847.5  5182.721     37 1.0000
## 317  495340   547287.52     492557.94      602017.1  8255.667     38 1.0000
## 320  797825   746569.02     681818.78      811319.3  3989.125     42 1.0000
## 322  355724   324224.29     274312.87      374135.7  5081.771     39 1.0000
## 330  575000   423213.55     367413.29      479013.8  4107.143     35 1.0000
## 331  976817   927080.87     805773.47     1048388.3  3907.268     33 1.0000
## 334  976817   866734.01     812592.36      920875.6  4440.077     37 1.0000
## 341  495340   372992.50     318335.42      427649.6  6191.750     33 1.2228
## 342  694227   662858.89     607400.83      718317.0  6942.270     33 2.0000
## 350  301133   275225.14     223896.67      326553.6  5018.883     33 1.0000
## 352 1286703  1587243.24    1478920.68     1695565.8  3216.758     39 1.0000
## 356  930000   892757.90     832821.41      952694.4  8086.957     43 2.0000
## 357  565400  1370993.92    1289163.29     1452824.5 14135.000     44 1.0000
## 368  732981   916881.99     856914.24      976849.7  5235.579     40 1.0000
## 377  580000  1059197.56     996243.85     1122151.3  1933.333     39 1.0000
## 381  955867   866896.41     784771.01      949021.8  6827.622     43 1.0000
## 384  772665   701151.18     655424.45      746877.9  5519.036     39 1.0000
## 385 1200000  1366090.66    1283232.64     1448948.7  9230.770     42 2.0000
## 393 1800000  1448019.78    1332230.78     1563808.8  3964.758     48 3.0000
## 395  500000   598359.19     554856.96      641861.4  5000.000     37 1.0000
## 397 1252320  1319567.95    1244003.56     1395132.3  3339.520     36 1.0000
##      nfull  npart   naux hoursw   hourspw      inv1     inv2 ssize
## 1   1.0000 1.0000 1.5357     76 16.755960  17166.67 27177.04   170
## 3   2.0000 2.2222 1.4091    114 17.191200 292857.20 71570.55   300
## 12  1.9556 1.0000 1.3673     40  7.514700  22207.04  5000.00    40
## 15  2.2656 2.0741 1.3333     92 13.786900  62269.23 16624.89    40
## 20  1.9556 1.2833 1.3673     65 11.594310  22207.04 22859.85    70
## 25  1.9556 1.2833 2.0000     99 15.868180  22207.04 22859.85    95
## 30  4.3590 2.0000 1.4091     84  9.580183 292857.20 71570.55   120
## 32  1.9231 1.5946 1.5357     86 12.192700  17166.67 27177.04   100
## 39  1.0000 2.2222 1.0000     96 18.383060 292857.20 71570.55   140
## 59  1.0000 2.0741 1.3333     88 13.734120  25000.00 16624.89   120
## 63  1.9556 1.0000 1.0000     78 15.739770  22207.04  5000.00    65
## 64  2.0000 1.5946 1.5357    117 19.085530  17166.67 27177.04    90
## 75  1.9556 1.2833 1.3673     43  7.670080  22207.04 22859.85    35
## 76  1.9556 1.2833 1.3673    104 18.550890  22207.04 22859.85   113
## 78  1.0000 1.2833 1.0000     63 14.708290  22207.04  1000.00   100
## 83  3.0000 2.0000 1.3673    154 20.289590  22207.04 22859.85    70
## 84  3.0000 2.0000 1.0000    148 20.490670  22207.04 15000.00    70
## 94  1.9556 1.2833 2.0000     92 14.746190  22207.04 22859.85   150
## 95  1.9556 2.0000 1.3673    104 16.448150  22207.04 22859.85   200
## 103 2.2656 2.0741 1.3333     72 10.789750  62269.23 16624.89   110
## 105 1.0000 1.0000 1.3333    126 29.077150 350000.00 16624.89   160
## 107 2.2656 1.0000 1.3333    108 16.366360  62269.23 16624.89   119
## 110 1.9556 2.0000 1.3673    125 17.069740 100000.00 10000.00   160
## 116 1.9556 1.2833 1.3673     32  5.707966   2500.00 22859.85    80
## 118 1.0000 1.2833 1.3673     72 15.481870  22207.04 22859.85   100
## 127 1.9556 1.2833 1.3673    100 15.137300  22207.04 22859.85   150
## 134 2.2656 2.0741 1.3333     70  9.122898 230000.00  3500.00   100
## 136 1.0000 2.2222 1.4091     92 16.337260 292857.20 71570.55    70
## 140 2.2656 2.0000 1.3333    100 15.154040   2000.00  2000.00   185
## 141 2.0000 2.0000 2.0000    175 25.000000 292857.20 25000.00   175
## 151 1.9231 1.5946 1.5357     86 14.206890  17166.67 27177.04    70
## 152 1.9556 1.2833 1.0000     80 15.270380  22207.04  6500.00   120
## 154 4.0000 1.2833 1.3673    240 27.743740  22207.04 30000.00   110
## 187 4.0000 3.0000 1.0000    313 34.777780  17166.67 72000.00   410
## 189 4.3590 2.2222 1.4091    150 16.684650 292857.20 71570.55    70
## 200 1.0000 3.0000 1.3333    102 16.105350  62269.23 16624.89   140
## 201 1.0000 2.0000 1.0000     92 18.400000  62269.23  5000.00   140
## 203 1.9556 1.2833 1.3673     90 16.053660  22207.04 22859.85    80
## 219 2.0000 2.0000 1.3333    170 26.842250  62269.23 16624.89   335
## 220 2.0000 1.5946 1.5357    160 22.439450  17166.67 27177.04   450
## 221 1.9556 1.0000 1.3673     63 11.835650  22207.04 22859.85   100
## 230 2.2656 2.0000 1.3333    126 16.581350  62269.23 16624.89   380
## 237 1.9556 1.2833 1.3673     87 15.518530  22207.04 22859.85   171
## 245 1.9556 1.2833 1.0000     82 15.652140   6000.00  5000.00    65
## 246 1.9556 3.0000 1.3673    160 19.224070  22207.04  5000.00   180
## 251 1.0000 2.0000 1.0000    171 28.500000 292857.20  1483.00   260
## 260 2.0000 1.0000 1.0000    165 27.500000 292857.20 71570.55   350
## 262 1.9556 1.2833 1.3673     62 11.059180  22207.04  1500.00    72
## 270 1.9556 1.2833 1.0000     55 10.498390  40000.00 48450.00    80
## 271 1.0000 1.0000 2.0000    173 28.833330  62269.23 16624.89   200
## 272 1.9556 1.2833 1.3673     87 15.518530  22207.04 22859.85   100
## 274 1.9556 1.0000 1.3673     90 16.908080  22207.04 22859.85   110
## 278 1.9556 1.2833 1.3673    100 15.137300  22207.04  5000.00    80
## 283 1.0000 1.0000 1.5357    192 34.683960  17166.67  2000.00   150
## 289 1.0000 2.2222 1.0000     86 16.468150 292857.20 71570.55   250
## 290 2.0000 3.0000 1.3333    186 25.363750  62269.23 16624.89   213
## 305 1.9231 1.5946 1.0000     65  9.972843  17166.67 10000.00   100
## 308 3.0000 1.2833 1.3673    162 24.358700  22207.04 22859.85   190
## 317 1.9556 1.2833 1.3673     52  9.275445  22207.04 22859.85    60
## 320 2.2656 1.0000 2.0000     94 15.002550  62269.23  1900.00   200
## 322 1.9556 1.2833 1.0000     57 10.880150  22207.04 22859.85    70
## 330 1.0000 2.0741 1.3333     70 12.945220  62269.23 16624.89   140
## 331 1.0000 1.5946 3.0000    122 18.499980  17166.67 27177.04   250
## 334 1.9231 1.5946 2.0000    113 17.337400  17166.67  5000.00   220
## 341 1.0000 1.0000 1.3673     55 11.982310  22207.04  5000.00    80
## 342 2.2656 2.0741 1.3333    104 13.554020  62269.23 16624.89   100
## 350 1.9556 1.2833 1.3673     45  8.026828  22207.04 22859.85    60
## 352 4.0000 1.0000 1.0000    228 32.571430  17166.67 27177.04   400
## 356 1.0000 1.2833 1.3673    150 26.545850  10000.00 22859.85   115
## 357 3.0000 2.0741 1.3333    167 22.545020  62269.23  5000.00    40
## 368 2.0000 1.2833 1.3673    163 28.846490  22207.04 22859.85   140
## 377 3.0000 2.0000 1.3333    176 24.000110  60000.00 30000.00   300
## 381 2.0000 3.0000 1.5357    122 16.189600   9000.00  7600.00   140
## 384 1.0000 1.2833 1.3673    105 22.577730  20000.00 22859.85   140
## 385 3.0000 1.0000 1.3673    250 33.933730  22207.04 22500.00   130
## 393 2.2656 3.0000 1.3333    221 23.023470  62269.23  2450.00   454
## 395 1.9231 1.0000 1.5357    104 19.051810  17166.67 80000.00   100
## 397 2.0000 1.5946 2.0000    170 25.778670  17166.67 27177.04   375

8 Kesimpulan

Kesimpulannya, model ini memiliki Adjusted R-squared : 0.795 sehingga dapat memprediksi 80%. Model ini juga masih kurang baik jika hendak digunakan untuk memprediksi harga penjualan busana terkait dataset ini dan untuk mendapatkan model lebih baik lagi maka dapat dilakukan dengan metode yang lain.