Regresi dengan pendekatan smoothing spline Pada Data Google Mobility Index dan Covid-19 di Jakarta Juli 2021
Daurin Nabilatul Munna, Prof. Dr. Suhartono, M.Kom
2022-06-03
1. Pengertian
Regresi Smoothing Spline merupakan salah satu metode nonparametrik bersifat piecewise polynomial, yaitu suatu potonganpotongan polinom yang memiliki sifat tersegmen pada selang k yang terbentuk pada titik-titik knot Wang and Yang (2009). Hal ini berarti fungsi Spline merupakan suatu gabungan beberapa polynomial pada knot-knot dengan suatu cara yang menjamin sifat kontuitas. Sehingga hasil prapemrosesan yang diperoleh dapat lebih akurat. Smoothing Spline memiliki kemampuan yang baik dalam mengidentifikasi bentuk kurva antara umur dan waktu bertahan hidup.
### 2. Data Google Mobility Index dan Covid-19 di Jakarta Juli 2021
library(readxl)mobilitycovidjakarta <- read_excel(path = "D:/matkul sem 2/linear/data covid jakarta 2021 mobile index.xlsx")
mobilitycovidjakarta ## # A tibble: 31 x 13
## Nama_provinsi Tanggal POSITIF Dirawat Sembuh Meninggal
## <chr> <dttm> <dbl> <dbl> <dbl> <dbl>
## 1 DKI Jakarta 2021-07-01 00:00:00 551009 24184 468461 8528
## 2 DKI Jakarta 2021-07-02 00:00:00 560408 25380 473467 8547
## 3 DKI Jakarta 2021-07-03 00:00:00 570110 27442 479150 8577
## 4 DKI Jakarta 2021-07-04 00:00:00 580595 27687 484949 8652
## 5 DKI Jakarta 2021-07-05 00:00:00 591498 28290 491556 8779
## 6 DKI Jakarta 2021-07-06 00:00:00 600937 29136 497492 8861
## 7 DKI Jakarta 2021-07-07 00:00:00 610303 30418 501199 9042
## 8 DKI Jakarta 2021-07-08 00:00:00 623277 29721 512085 9110
## 9 DKI Jakarta 2021-07-09 00:00:00 636389 26903 526941 9306
## 10 DKI Jakarta 2021-07-10 00:00:00 649309 24273 543867 9357
## # ... with 21 more rows, and 7 more variables: Self_isolation <dbl>,
## # retail_and_recreation_percent_change_from_baseline <dbl>,
## # grocery_and_pharmacy_percent_change_from_baseline <dbl>,
## # parks_percent_change_from_baseline <dbl>,
## # transit_stations_percent_change_from_baseline <dbl>,
## # workplaces_percent_change_from_baseline <dbl>,
## # residential_percent_change_from_baseline <dbl>a. Positif
plot(mobilitycovidjakarta$Tanggal,mobilitycovidjakarta$POSITIF)
b. Dirawat
plot(mobilitycovidjakarta$Tanggal,mobilitycovidjakarta$Dirawat)
c. Sembuh
plot(mobilitycovidjakarta$Tanggal,mobilitycovidjakarta$Sembuh)
d. Meninggal
plot(mobilitycovidjakarta$Tanggal,mobilitycovidjakarta$Meninggal)
e. Self Isolation
plot(mobilitycovidjakarta$Tanggal,mobilitycovidjakarta$`Self_isolation`)
3. Data Deskriptif
library(jmv)a. Positif
# Mendapatkan data descriptive menggunakan fungsi descritptive
descriptives(mobilitycovidjakarta, vars = vars(POSITIF, grocery_and_pharmacy_percent_change_from_baseline), freq = TRUE)##
## DESCRIPTIVES
##
## Descriptives
## ---------------------------------------------------------------------------------------
## POSITIF grocery_and_pharmacy_percent_change_from_baseline
## ---------------------------------------------------------------------------------------
## N 31 31
## Missing 0 0
## Mean 705204.7 -2.161290
## Median 727016.0 -2.000000
## Standard deviation 86507.77 4.058709
## Minimum 551009.0 -13.00000
## Maximum 814653.0 6.000000
## ---------------------------------------------------------------------------------------b. Dirawat
# Mendapatkan data descriptive menggunakan fungsi descritptive
descriptives(mobilitycovidjakarta, vars = vars(Dirawat, grocery_and_pharmacy_percent_change_from_baseline), freq = TRUE)##
## DESCRIPTIVES
##
## Descriptives
## ---------------------------------------------------------------------------------------
## Dirawat grocery_and_pharmacy_percent_change_from_baseline
## ---------------------------------------------------------------------------------------
## N 31 31
## Missing 0 0
## Mean 20398.68 -2.161290
## Median 20359.00 -2.000000
## Standard deviation 6343.437 4.058709
## Minimum 7716.000 -13.00000
## Maximum 30418.00 6.000000
## ---------------------------------------------------------------------------------------c. Sembuh
# Mendapatkan data descriptive menggunakan fungsi descritptive
descriptives(mobilitycovidjakarta, vars = vars(Sembuh, grocery_and_pharmacy_percent_change_from_baseline), freq = TRUE)##
## DESCRIPTIVES
##
## Descriptives
## ---------------------------------------------------------------------------------------
## Sembuh grocery_and_pharmacy_percent_change_from_baseline
## ---------------------------------------------------------------------------------------
## N 31 31
## Missing 0 0
## Mean 614735.5 -2.161290
## Median 604060.0 -2.000000
## Standard deviation 101497.5 4.058709
## Minimum 468461.0 -13.00000
## Maximum 784668.0 6.000000
## ---------------------------------------------------------------------------------------d. Meninggal
# Mendapatkan data descriptive menggunakan fungsi descritptive
descriptives(mobilitycovidjakarta, vars = vars(Meninggal, grocery_and_pharmacy_percent_change_from_baseline), freq = TRUE)##
## DESCRIPTIVES
##
## Descriptives
## ----------------------------------------------------------------------------------------
## Meninggal grocery_and_pharmacy_percent_change_from_baseline
## ----------------------------------------------------------------------------------------
## N 31 31
## Missing 0 0
## Mean 10097.87 -2.161290
## Median 9818.000 -2.000000
## Standard deviation 1148.204 4.058709
## Minimum 8528.000 -13.00000
## Maximum 12135.00 6.000000
## ----------------------------------------------------------------------------------------e. Self Isolation
# Mendapatkan data descriptive menggunakan fungsi descritptive
descriptives(mobilitycovidjakarta, vars = vars(Self_isolation, grocery_and_pharmacy_percent_change_from_baseline), freq = TRUE)##
## DESCRIPTIVES
##
## Descriptives
## ---------------------------------------------------------------------------------------------
## Self_isolation grocery_and_pharmacy_percent_change_from_baseline
## ---------------------------------------------------------------------------------------------
## N 31 31
## Missing 0 0
## Mean 59972.65 -2.161290
## Median 66581.00 -2.000000
## Standard deviation 22461.30 4.058709
## Minimum 10134.00 -13.00000
## Maximum 88295.00 6.000000
## ---------------------------------------------------------------------------------------------4. Perbandingan Pendekatan
library(npreg)a. Positif
mod.ss <- ss(mobilitycovidjakarta$Tanggal,mobilitycovidjakarta$POSITIF, nknots = 10)
mobilitycovidjakarta$prediksi_ss <- mod.ss$y
mobilitycovidjakarta## # A tibble: 31 x 14
## Nama_provinsi Tanggal POSITIF Dirawat Sembuh Meninggal
## <chr> <dttm> <dbl> <dbl> <dbl> <dbl>
## 1 DKI Jakarta 2021-07-01 00:00:00 551009 24184 468461 8528
## 2 DKI Jakarta 2021-07-02 00:00:00 560408 25380 473467 8547
## 3 DKI Jakarta 2021-07-03 00:00:00 570110 27442 479150 8577
## 4 DKI Jakarta 2021-07-04 00:00:00 580595 27687 484949 8652
## 5 DKI Jakarta 2021-07-05 00:00:00 591498 28290 491556 8779
## 6 DKI Jakarta 2021-07-06 00:00:00 600937 29136 497492 8861
## 7 DKI Jakarta 2021-07-07 00:00:00 610303 30418 501199 9042
## 8 DKI Jakarta 2021-07-08 00:00:00 623277 29721 512085 9110
## 9 DKI Jakarta 2021-07-09 00:00:00 636389 26903 526941 9306
## 10 DKI Jakarta 2021-07-10 00:00:00 649309 24273 543867 9357
## # ... with 21 more rows, and 8 more variables: Self_isolation <dbl>,
## # retail_and_recreation_percent_change_from_baseline <dbl>,
## # grocery_and_pharmacy_percent_change_from_baseline <dbl>,
## # parks_percent_change_from_baseline <dbl>,
## # transit_stations_percent_change_from_baseline <dbl>,
## # workplaces_percent_change_from_baseline <dbl>,
## # residential_percent_change_from_baseline <dbl>, prediksi_ss <dbl>b. Dirawat
mod.ss <- ss(mobilitycovidjakarta$Tanggal,mobilitycovidjakarta$Dirawat, nknots = 10)
mobilitycovidjakarta$prediksi_ss <- mod.ss$y
mobilitycovidjakarta## # A tibble: 31 x 14
## Nama_provinsi Tanggal POSITIF Dirawat Sembuh Meninggal
## <chr> <dttm> <dbl> <dbl> <dbl> <dbl>
## 1 DKI Jakarta 2021-07-01 00:00:00 551009 24184 468461 8528
## 2 DKI Jakarta 2021-07-02 00:00:00 560408 25380 473467 8547
## 3 DKI Jakarta 2021-07-03 00:00:00 570110 27442 479150 8577
## 4 DKI Jakarta 2021-07-04 00:00:00 580595 27687 484949 8652
## 5 DKI Jakarta 2021-07-05 00:00:00 591498 28290 491556 8779
## 6 DKI Jakarta 2021-07-06 00:00:00 600937 29136 497492 8861
## 7 DKI Jakarta 2021-07-07 00:00:00 610303 30418 501199 9042
## 8 DKI Jakarta 2021-07-08 00:00:00 623277 29721 512085 9110
## 9 DKI Jakarta 2021-07-09 00:00:00 636389 26903 526941 9306
## 10 DKI Jakarta 2021-07-10 00:00:00 649309 24273 543867 9357
## # ... with 21 more rows, and 8 more variables: Self_isolation <dbl>,
## # retail_and_recreation_percent_change_from_baseline <dbl>,
## # grocery_and_pharmacy_percent_change_from_baseline <dbl>,
## # parks_percent_change_from_baseline <dbl>,
## # transit_stations_percent_change_from_baseline <dbl>,
## # workplaces_percent_change_from_baseline <dbl>,
## # residential_percent_change_from_baseline <dbl>, prediksi_ss <dbl>c. Sembuh
mod.ss <- ss(mobilitycovidjakarta$Tanggal,mobilitycovidjakarta$Sembuh, nknots = 10)
mobilitycovidjakarta$prediksi_ss <- mod.ss$y
mobilitycovidjakarta## # A tibble: 31 x 14
## Nama_provinsi Tanggal POSITIF Dirawat Sembuh Meninggal
## <chr> <dttm> <dbl> <dbl> <dbl> <dbl>
## 1 DKI Jakarta 2021-07-01 00:00:00 551009 24184 468461 8528
## 2 DKI Jakarta 2021-07-02 00:00:00 560408 25380 473467 8547
## 3 DKI Jakarta 2021-07-03 00:00:00 570110 27442 479150 8577
## 4 DKI Jakarta 2021-07-04 00:00:00 580595 27687 484949 8652
## 5 DKI Jakarta 2021-07-05 00:00:00 591498 28290 491556 8779
## 6 DKI Jakarta 2021-07-06 00:00:00 600937 29136 497492 8861
## 7 DKI Jakarta 2021-07-07 00:00:00 610303 30418 501199 9042
## 8 DKI Jakarta 2021-07-08 00:00:00 623277 29721 512085 9110
## 9 DKI Jakarta 2021-07-09 00:00:00 636389 26903 526941 9306
## 10 DKI Jakarta 2021-07-10 00:00:00 649309 24273 543867 9357
## # ... with 21 more rows, and 8 more variables: Self_isolation <dbl>,
## # retail_and_recreation_percent_change_from_baseline <dbl>,
## # grocery_and_pharmacy_percent_change_from_baseline <dbl>,
## # parks_percent_change_from_baseline <dbl>,
## # transit_stations_percent_change_from_baseline <dbl>,
## # workplaces_percent_change_from_baseline <dbl>,
## # residential_percent_change_from_baseline <dbl>, prediksi_ss <dbl>d. Meninggal
mod.ss <- ss(mobilitycovidjakarta$Tanggal,mobilitycovidjakarta$Meninggal, nknots = 10)
mobilitycovidjakarta$prediksi_ss <- mod.ss$y
mobilitycovidjakarta## # A tibble: 31 x 14
## Nama_provinsi Tanggal POSITIF Dirawat Sembuh Meninggal
## <chr> <dttm> <dbl> <dbl> <dbl> <dbl>
## 1 DKI Jakarta 2021-07-01 00:00:00 551009 24184 468461 8528
## 2 DKI Jakarta 2021-07-02 00:00:00 560408 25380 473467 8547
## 3 DKI Jakarta 2021-07-03 00:00:00 570110 27442 479150 8577
## 4 DKI Jakarta 2021-07-04 00:00:00 580595 27687 484949 8652
## 5 DKI Jakarta 2021-07-05 00:00:00 591498 28290 491556 8779
## 6 DKI Jakarta 2021-07-06 00:00:00 600937 29136 497492 8861
## 7 DKI Jakarta 2021-07-07 00:00:00 610303 30418 501199 9042
## 8 DKI Jakarta 2021-07-08 00:00:00 623277 29721 512085 9110
## 9 DKI Jakarta 2021-07-09 00:00:00 636389 26903 526941 9306
## 10 DKI Jakarta 2021-07-10 00:00:00 649309 24273 543867 9357
## # ... with 21 more rows, and 8 more variables: Self_isolation <dbl>,
## # retail_and_recreation_percent_change_from_baseline <dbl>,
## # grocery_and_pharmacy_percent_change_from_baseline <dbl>,
## # parks_percent_change_from_baseline <dbl>,
## # transit_stations_percent_change_from_baseline <dbl>,
## # workplaces_percent_change_from_baseline <dbl>,
## # residential_percent_change_from_baseline <dbl>, prediksi_ss <dbl>e. Self Isolation
mod.ss <- ss(mobilitycovidjakarta$Tanggal,mobilitycovidjakarta$Self_isolation, nknots = 10)
mobilitycovidjakarta$prediksi_ss <- mod.ss$y
mobilitycovidjakarta## # A tibble: 31 x 14
## Nama_provinsi Tanggal POSITIF Dirawat Sembuh Meninggal
## <chr> <dttm> <dbl> <dbl> <dbl> <dbl>
## 1 DKI Jakarta 2021-07-01 00:00:00 551009 24184 468461 8528
## 2 DKI Jakarta 2021-07-02 00:00:00 560408 25380 473467 8547
## 3 DKI Jakarta 2021-07-03 00:00:00 570110 27442 479150 8577
## 4 DKI Jakarta 2021-07-04 00:00:00 580595 27687 484949 8652
## 5 DKI Jakarta 2021-07-05 00:00:00 591498 28290 491556 8779
## 6 DKI Jakarta 2021-07-06 00:00:00 600937 29136 497492 8861
## 7 DKI Jakarta 2021-07-07 00:00:00 610303 30418 501199 9042
## 8 DKI Jakarta 2021-07-08 00:00:00 623277 29721 512085 9110
## 9 DKI Jakarta 2021-07-09 00:00:00 636389 26903 526941 9306
## 10 DKI Jakarta 2021-07-10 00:00:00 649309 24273 543867 9357
## # ... with 21 more rows, and 8 more variables: Self_isolation <dbl>,
## # retail_and_recreation_percent_change_from_baseline <dbl>,
## # grocery_and_pharmacy_percent_change_from_baseline <dbl>,
## # parks_percent_change_from_baseline <dbl>,
## # transit_stations_percent_change_from_baseline <dbl>,
## # workplaces_percent_change_from_baseline <dbl>,
## # residential_percent_change_from_baseline <dbl>, prediksi_ss <dbl>4. Perbandingan Pendekatan
library(npreg)a. Positif
mod.ss <- ss(mobilitycovidjakarta$Tanggal,mobilitycovidjakarta$POSITIF, nknots = 10)
mobilitycovidjakarta$prediksi_ss <- mod.ss$y
mobilitycovidjakarta## # A tibble: 31 x 14
## Nama_provinsi Tanggal POSITIF Dirawat Sembuh Meninggal
## <chr> <dttm> <dbl> <dbl> <dbl> <dbl>
## 1 DKI Jakarta 2021-07-01 00:00:00 551009 24184 468461 8528
## 2 DKI Jakarta 2021-07-02 00:00:00 560408 25380 473467 8547
## 3 DKI Jakarta 2021-07-03 00:00:00 570110 27442 479150 8577
## 4 DKI Jakarta 2021-07-04 00:00:00 580595 27687 484949 8652
## 5 DKI Jakarta 2021-07-05 00:00:00 591498 28290 491556 8779
## 6 DKI Jakarta 2021-07-06 00:00:00 600937 29136 497492 8861
## 7 DKI Jakarta 2021-07-07 00:00:00 610303 30418 501199 9042
## 8 DKI Jakarta 2021-07-08 00:00:00 623277 29721 512085 9110
## 9 DKI Jakarta 2021-07-09 00:00:00 636389 26903 526941 9306
## 10 DKI Jakarta 2021-07-10 00:00:00 649309 24273 543867 9357
## # ... with 21 more rows, and 8 more variables: Self_isolation <dbl>,
## # retail_and_recreation_percent_change_from_baseline <dbl>,
## # grocery_and_pharmacy_percent_change_from_baseline <dbl>,
## # parks_percent_change_from_baseline <dbl>,
## # transit_stations_percent_change_from_baseline <dbl>,
## # workplaces_percent_change_from_baseline <dbl>,
## # residential_percent_change_from_baseline <dbl>, prediksi_ss <dbl>b. Dirawat
mod.ss <- ss(mobilitycovidjakarta$Tanggal,mobilitycovidjakarta$Dirawat, nknots = 10)
mobilitycovidjakarta$prediksi_ss <- mod.ss$y
mobilitycovidjakarta## # A tibble: 31 x 14
## Nama_provinsi Tanggal POSITIF Dirawat Sembuh Meninggal
## <chr> <dttm> <dbl> <dbl> <dbl> <dbl>
## 1 DKI Jakarta 2021-07-01 00:00:00 551009 24184 468461 8528
## 2 DKI Jakarta 2021-07-02 00:00:00 560408 25380 473467 8547
## 3 DKI Jakarta 2021-07-03 00:00:00 570110 27442 479150 8577
## 4 DKI Jakarta 2021-07-04 00:00:00 580595 27687 484949 8652
## 5 DKI Jakarta 2021-07-05 00:00:00 591498 28290 491556 8779
## 6 DKI Jakarta 2021-07-06 00:00:00 600937 29136 497492 8861
## 7 DKI Jakarta 2021-07-07 00:00:00 610303 30418 501199 9042
## 8 DKI Jakarta 2021-07-08 00:00:00 623277 29721 512085 9110
## 9 DKI Jakarta 2021-07-09 00:00:00 636389 26903 526941 9306
## 10 DKI Jakarta 2021-07-10 00:00:00 649309 24273 543867 9357
## # ... with 21 more rows, and 8 more variables: Self_isolation <dbl>,
## # retail_and_recreation_percent_change_from_baseline <dbl>,
## # grocery_and_pharmacy_percent_change_from_baseline <dbl>,
## # parks_percent_change_from_baseline <dbl>,
## # transit_stations_percent_change_from_baseline <dbl>,
## # workplaces_percent_change_from_baseline <dbl>,
## # residential_percent_change_from_baseline <dbl>, prediksi_ss <dbl>c. Sembuh
mod.ss <- ss(mobilitycovidjakarta$Tanggal,mobilitycovidjakarta$Sembuh, nknots = 10)
mobilitycovidjakarta$prediksi_ss <- mod.ss$y
mobilitycovidjakarta## # A tibble: 31 x 14
## Nama_provinsi Tanggal POSITIF Dirawat Sembuh Meninggal
## <chr> <dttm> <dbl> <dbl> <dbl> <dbl>
## 1 DKI Jakarta 2021-07-01 00:00:00 551009 24184 468461 8528
## 2 DKI Jakarta 2021-07-02 00:00:00 560408 25380 473467 8547
## 3 DKI Jakarta 2021-07-03 00:00:00 570110 27442 479150 8577
## 4 DKI Jakarta 2021-07-04 00:00:00 580595 27687 484949 8652
## 5 DKI Jakarta 2021-07-05 00:00:00 591498 28290 491556 8779
## 6 DKI Jakarta 2021-07-06 00:00:00 600937 29136 497492 8861
## 7 DKI Jakarta 2021-07-07 00:00:00 610303 30418 501199 9042
## 8 DKI Jakarta 2021-07-08 00:00:00 623277 29721 512085 9110
## 9 DKI Jakarta 2021-07-09 00:00:00 636389 26903 526941 9306
## 10 DKI Jakarta 2021-07-10 00:00:00 649309 24273 543867 9357
## # ... with 21 more rows, and 8 more variables: Self_isolation <dbl>,
## # retail_and_recreation_percent_change_from_baseline <dbl>,
## # grocery_and_pharmacy_percent_change_from_baseline <dbl>,
## # parks_percent_change_from_baseline <dbl>,
## # transit_stations_percent_change_from_baseline <dbl>,
## # workplaces_percent_change_from_baseline <dbl>,
## # residential_percent_change_from_baseline <dbl>, prediksi_ss <dbl>d. Meninggal
mod.ss <- ss(mobilitycovidjakarta$Tanggal,mobilitycovidjakarta$Meninggal, nknots = 10)
mobilitycovidjakarta$prediksi_ss <- mod.ss$y
mobilitycovidjakarta## # A tibble: 31 x 14
## Nama_provinsi Tanggal POSITIF Dirawat Sembuh Meninggal
## <chr> <dttm> <dbl> <dbl> <dbl> <dbl>
## 1 DKI Jakarta 2021-07-01 00:00:00 551009 24184 468461 8528
## 2 DKI Jakarta 2021-07-02 00:00:00 560408 25380 473467 8547
## 3 DKI Jakarta 2021-07-03 00:00:00 570110 27442 479150 8577
## 4 DKI Jakarta 2021-07-04 00:00:00 580595 27687 484949 8652
## 5 DKI Jakarta 2021-07-05 00:00:00 591498 28290 491556 8779
## 6 DKI Jakarta 2021-07-06 00:00:00 600937 29136 497492 8861
## 7 DKI Jakarta 2021-07-07 00:00:00 610303 30418 501199 9042
## 8 DKI Jakarta 2021-07-08 00:00:00 623277 29721 512085 9110
## 9 DKI Jakarta 2021-07-09 00:00:00 636389 26903 526941 9306
## 10 DKI Jakarta 2021-07-10 00:00:00 649309 24273 543867 9357
## # ... with 21 more rows, and 8 more variables: Self_isolation <dbl>,
## # retail_and_recreation_percent_change_from_baseline <dbl>,
## # grocery_and_pharmacy_percent_change_from_baseline <dbl>,
## # parks_percent_change_from_baseline <dbl>,
## # transit_stations_percent_change_from_baseline <dbl>,
## # workplaces_percent_change_from_baseline <dbl>,
## # residential_percent_change_from_baseline <dbl>, prediksi_ss <dbl>e. Self Isolation
mod.ss <- ss(mobilitycovidjakarta$Tanggal,mobilitycovidjakarta$Self_isolation, nknots = 10)
mobilitycovidjakarta$prediksi_ss <- mod.ss$y
mobilitycovidjakarta## # A tibble: 31 x 14
## Nama_provinsi Tanggal POSITIF Dirawat Sembuh Meninggal
## <chr> <dttm> <dbl> <dbl> <dbl> <dbl>
## 1 DKI Jakarta 2021-07-01 00:00:00 551009 24184 468461 8528
## 2 DKI Jakarta 2021-07-02 00:00:00 560408 25380 473467 8547
## 3 DKI Jakarta 2021-07-03 00:00:00 570110 27442 479150 8577
## 4 DKI Jakarta 2021-07-04 00:00:00 580595 27687 484949 8652
## 5 DKI Jakarta 2021-07-05 00:00:00 591498 28290 491556 8779
## 6 DKI Jakarta 2021-07-06 00:00:00 600937 29136 497492 8861
## 7 DKI Jakarta 2021-07-07 00:00:00 610303 30418 501199 9042
## 8 DKI Jakarta 2021-07-08 00:00:00 623277 29721 512085 9110
## 9 DKI Jakarta 2021-07-09 00:00:00 636389 26903 526941 9306
## 10 DKI Jakarta 2021-07-10 00:00:00 649309 24273 543867 9357
## # ... with 21 more rows, and 8 more variables: Self_isolation <dbl>,
## # retail_and_recreation_percent_change_from_baseline <dbl>,
## # grocery_and_pharmacy_percent_change_from_baseline <dbl>,
## # parks_percent_change_from_baseline <dbl>,
## # transit_stations_percent_change_from_baseline <dbl>,
## # workplaces_percent_change_from_baseline <dbl>,
## # residential_percent_change_from_baseline <dbl>, prediksi_ss <dbl># Hasil plot
plot(mobilitycovidjakarta$Tanggal, mobilitycovidjakarta$workplaces_percent_change_from_baseline, lty = 10, col = 'black', lwd =5)
# add lm fit
abline(coef(lm( mobilitycovidjakarta$workplaces_percent_change_from_baseline ~ mobilitycovidjakarta$Tanggal , data = mobilitycovidjakarta)), lty = 3, col = 'green', lwd =4)
lines(mobilitycovidjakarta$Tanggal, mobilitycovidjakarta$prediksi_ss1 , lty = 2, col = 'Blue', lwd = 4)## Warning: Unknown or uninitialised column: `prediksi_ss1`.lines(mobilitycovidjakarta$Tanggal, mobilitycovidjakarta$prediksi_smsp1, lty = 4, col = 'Orange', lwd = 6)## Warning: Unknown or uninitialised column: `prediksi_smsp1`.
5. Korelasi Pearson
Korelasi Pearson adalah suatu alat yang meimiliki fungsi menganalisis statistik yang digunakan untuk melihat keeratan hubungan linier antara 2 variabel yang skala datanya adalah interval atau rasio.
a, Positif
cor.test(mobilitycovidjakarta$POSITIF,mobilitycovidjakarta$workplaces_percent_change_from_baseline)##
## Pearson's product-moment correlation
##
## data: mobilitycovidjakarta$POSITIF and mobilitycovidjakarta$workplaces_percent_change_from_baseline
## t = 1.0596, df = 29, p-value = 0.2981
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.1731184 0.5123545
## sample estimates:
## cor
## 0.1930649b. Sembuh
cor.test(mobilitycovidjakarta$Sembuh,mobilitycovidjakarta$workplaces_percent_change_from_baseline)##
## Pearson's product-moment correlation
##
## data: mobilitycovidjakarta$Sembuh and mobilitycovidjakarta$workplaces_percent_change_from_baseline
## t = 1.2263, df = 29, p-value = 0.23
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.1436114 0.5343298
## sample estimates:
## cor
## 0.2220271c. Dirawat
cor.test(mobilitycovidjakarta$Dirawat,mobilitycovidjakarta$workplaces_percent_change_from_baseline)##
## Pearson's product-moment correlation
##
## data: mobilitycovidjakarta$Dirawat and mobilitycovidjakarta$workplaces_percent_change_from_baseline
## t = -1.0773, df = 29, p-value = 0.2902
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.5147307 0.1699860
## sample estimates:
## cor
## -0.19617d. Meninggal
cor.test(mobilitycovidjakarta$Meninggal,mobilitycovidjakarta$workplaces_percent_change_from_baseline)##
## Pearson's product-moment correlation
##
## data: mobilitycovidjakarta$Meninggal and mobilitycovidjakarta$workplaces_percent_change_from_baseline
## t = 1.4017, df = 29, p-value = 0.1716
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.1124889 0.5565560
## sample estimates:
## cor
## 0.2518914e. Self Isolation
cor.test(mobilitycovidjakarta$Self_isolation,mobilitycovidjakarta$workplaces_percent_change_from_baseline)##
## Pearson's product-moment correlation
##
## data: mobilitycovidjakarta$Self_isolation and mobilitycovidjakarta$workplaces_percent_change_from_baseline
## t = -1.1982, df = 29, p-value = 0.2405
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.5306901 0.1485833
## sample estimates:
## cor
## -0.2171917Dari output yang telah terbuat dapat disimpulkan bahwa tidak adanya hubungan yang signifikan antara mobilityjakarta Dirawat dan mobilityjakarta workplaces_percent_change_from_baseline (p-value<0,01). Nilai p-value dalam output R dituliskan 0.01321. Nilai koefisien korelasi r adalah sebesar 0.4401926yang menunjukkan hubungan yang lemah dan positif ( berbanding lurus) antara variabel Dirawat dan workplaces_percent_change_from_baseline.
6. Hasil Regresi dengan Pendekatan Smoothing Spline
a. Positif
model <- lm(mobilitycovidjakarta$POSITIF ~ mobilitycovidjakarta$workplaces_percent_change_from_baseline, data = mobilitycovidjakarta)
summary(model)##
## Call:
## lm(formula = mobilitycovidjakarta$POSITIF ~ mobilitycovidjakarta$workplaces_percent_change_from_baseline,
## data = mobilitycovidjakarta)
##
## Residuals:
## Min 1Q Median 3Q Max
## -169377 -79643 28079 77994 110669
##
## Coefficients:
## Estimate
## (Intercept) 748143
## mobilitycovidjakarta$workplaces_percent_change_from_baseline 1262
## Std. Error t value
## (Intercept) 43387 17.24
## mobilitycovidjakarta$workplaces_percent_change_from_baseline 1191 1.06
## Pr(>|t|)
## (Intercept) <2e-16 ***
## mobilitycovidjakarta$workplaces_percent_change_from_baseline 0.298
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 86330 on 29 degrees of freedom
## Multiple R-squared: 0.03727, Adjusted R-squared: 0.004077
## F-statistic: 1.123 on 1 and 29 DF, p-value: 0.2981b. Dirawat
model <- lm(mobilitycovidjakarta$Dirawat ~ mobilitycovidjakarta$workplaces_percent_change_from_baseline, data = mobilitycovidjakarta)
summary(model)##
## Call:
## lm(formula = mobilitycovidjakarta$Dirawat ~ mobilitycovidjakarta$workplaces_percent_change_from_baseline,
## data = mobilitycovidjakarta)
##
## Residuals:
## Min 1Q Median 3Q Max
## -12774 -4487 -1063 4621 10868
##
## Coefficients:
## Estimate
## (Intercept) 17199.48
## mobilitycovidjakarta$workplaces_percent_change_from_baseline -94.00
## Std. Error t value
## (Intercept) 3179.50 5.409
## mobilitycovidjakarta$workplaces_percent_change_from_baseline 87.26 -1.077
## Pr(>|t|)
## (Intercept) 8.13e-06 ***
## mobilitycovidjakarta$workplaces_percent_change_from_baseline 0.29
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 6327 on 29 degrees of freedom
## Multiple R-squared: 0.03848, Adjusted R-squared: 0.005327
## F-statistic: 1.161 on 1 and 29 DF, p-value: 0.2902c. Sembuh
model <- lm(mobilitycovidjakarta$Sembuh ~ mobilitycovidjakarta$workplaces_percent_change_from_baseline, data = mobilitycovidjakarta)
summary(model)##
## Call:
## lm(formula = mobilitycovidjakarta$Sembuh ~ mobilitycovidjakarta$workplaces_percent_change_from_baseline,
## data = mobilitycovidjakarta)
##
## Residuals:
## Min 1Q Median 3Q Max
## -166758 -93211 5825 83785 171580
##
## Coefficients:
## Estimate
## (Intercept) 672671
## mobilitycovidjakarta$workplaces_percent_change_from_baseline 1702
## Std. Error t value
## (Intercept) 50586 13.297
## mobilitycovidjakarta$workplaces_percent_change_from_baseline 1388 1.226
## Pr(>|t|)
## (Intercept) 7.19e-14 ***
## mobilitycovidjakarta$workplaces_percent_change_from_baseline 0.23
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 100700 on 29 degrees of freedom
## Multiple R-squared: 0.0493, Adjusted R-squared: 0.01651
## F-statistic: 1.504 on 1 and 29 DF, p-value: 0.23d. Meninggal
model <- lm(mobilitycovidjakarta$Meninggal ~ mobilitycovidjakarta$workplaces_percent_change_from_baseline, data = mobilitycovidjakarta)
summary(model)##
## Call:
## lm(formula = mobilitycovidjakarta$Meninggal ~ mobilitycovidjakarta$workplaces_percent_change_from_baseline,
## data = mobilitycovidjakarta)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1832.76 -964.89 17.43 1046.12 2058.27
##
## Coefficients:
## Estimate
## (Intercept) 10841.43
## mobilitycovidjakarta$workplaces_percent_change_from_baseline 21.85
## Std. Error t value
## (Intercept) 567.99 19.087
## mobilitycovidjakarta$workplaces_percent_change_from_baseline 15.59 1.402
## Pr(>|t|)
## (Intercept) <2e-16 ***
## mobilitycovidjakarta$workplaces_percent_change_from_baseline 0.172
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1130 on 29 degrees of freedom
## Multiple R-squared: 0.06345, Adjusted R-squared: 0.03115
## F-statistic: 1.965 on 1 and 29 DF, p-value: 0.1716e. Self Isolation
model <- lm(mobilitycovidjakarta$Self_isolation ~ mobilitycovidjakarta$workplaces_percent_change_from_baseline, data = mobilitycovidjakarta)
summary(model)##
## Call:
## lm(formula = mobilitycovidjakarta$Self_isolation ~ mobilitycovidjakarta$workplaces_percent_change_from_baseline,
## data = mobilitycovidjakarta)
##
## Residuals:
## Min 1Q Median 3Q Max
## -50195 -8747 3645 16100 32971
##
## Coefficients:
## Estimate
## (Intercept) 47430.8
## mobilitycovidjakarta$workplaces_percent_change_from_baseline -368.5
## Std. Error t value
## (Intercept) 11207.2 4.232
## mobilitycovidjakarta$workplaces_percent_change_from_baseline 307.6 -1.198
## Pr(>|t|)
## (Intercept) 0.000212 ***
## mobilitycovidjakarta$workplaces_percent_change_from_baseline 0.240534
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 22300 on 29 degrees of freedom
## Multiple R-squared: 0.04717, Adjusted R-squared: 0.01432
## F-statistic: 1.436 on 1 and 29 DF, p-value: 0.2405mobilitycovidjakarta$prediksi_model <- model$fitted.values
mobilitycovidjakarta## # A tibble: 31 x 15
## Nama_provinsi Tanggal POSITIF Dirawat Sembuh Meninggal
## <chr> <dttm> <dbl> <dbl> <dbl> <dbl>
## 1 DKI Jakarta 2021-07-01 00:00:00 551009 24184 468461 8528
## 2 DKI Jakarta 2021-07-02 00:00:00 560408 25380 473467 8547
## 3 DKI Jakarta 2021-07-03 00:00:00 570110 27442 479150 8577
## 4 DKI Jakarta 2021-07-04 00:00:00 580595 27687 484949 8652
## 5 DKI Jakarta 2021-07-05 00:00:00 591498 28290 491556 8779
## 6 DKI Jakarta 2021-07-06 00:00:00 600937 29136 497492 8861
## 7 DKI Jakarta 2021-07-07 00:00:00 610303 30418 501199 9042
## 8 DKI Jakarta 2021-07-08 00:00:00 623277 29721 512085 9110
## 9 DKI Jakarta 2021-07-09 00:00:00 636389 26903 526941 9306
## 10 DKI Jakarta 2021-07-10 00:00:00 649309 24273 543867 9357
## # ... with 21 more rows, and 9 more variables: Self_isolation <dbl>,
## # retail_and_recreation_percent_change_from_baseline <dbl>,
## # grocery_and_pharmacy_percent_change_from_baseline <dbl>,
## # parks_percent_change_from_baseline <dbl>,
## # transit_stations_percent_change_from_baseline <dbl>,
## # workplaces_percent_change_from_baseline <dbl>,
## # residential_percent_change_from_baseline <dbl>, prediksi_ss <dbl>, ...Menambahkan Histogram
a. Positif
# Menambahkan Histograms
panel.hist <- function(x, ...) {
usr <- par("usr")
on.exit(par(usr))
par(usr = c(usr[1:2], 0, 1.5))
his <- hist(x, plot = FALSE)
breaks <- his$breaks
nB <- length(breaks)
y <- his$counts
y <- y/max(y)
rect(breaks[-nB], 0, breaks[-1], y, col = rgb(0, 0, 1, alpha = 0.5), ...)
# lines(density(x), col = 2, lwd = 2) # Uncomment to add density lines
}
# Menyetarakan berdasarkan formula
pairs(mobilitycovidjakarta$POSITIF~mobilitycovidjakarta$workplaces_percent_change_from_baseline, data = mobilitycovidjakarta,
upper.panel = NULL, # Disabling the upper panel
diag.panel = panel.hist) # Adding the histograms## Warning in par(usr): argument 1 does not name a graphical parameter
## Warning in par(usr): argument 1 does not name a graphical parameter
# plot method
plot(mobilitycovidjakarta$Tanggal, mobilitycovidjakarta$POSITIF, lty = 10, col = 'black', lwd =5)
#plot(mod.ss)
# add lm fit
abline(coef(lm( mobilitycovidjakarta$POSITIF ~ mobilitycovidjakarta$Tanggal , data = mobilitycovidjakarta)), lty = 10, col = 'purple', lwd =5)
rug(mobilitycovidjakarta$Tanggal) # add rug to plot
lines(mobilitycovidjakarta$Tanggal,mod.ss$y , lty = 2, col = 'blue', lwd = 4)
#plot(mod.smsp)
lines(mobilitycovidjakarta$Tanggal, mobilitycovidjakarta$prediksi_model, lty = 2, col = 'cyan', lwd = 4)
legend("topleft",
legend = c("Real", "Model SS", "Trends", "Regresi Nonparametrik"),
lty = 1:4, col = c("Black","blue","purple","cyan"), lwd = 3, bty = "p")
b. Dirawat
# Menambahkan Histograms
panel.hist <- function(x, ...) {
usr <- par("usr")
on.exit(par(usr))
par(usr = c(usr[1:2], 0, 1.5))
his <- hist(x, plot = FALSE)
breaks <- his$breaks
nB <- length(breaks)
y <- his$counts
y <- y/max(y)
rect(breaks[-nB], 0, breaks[-1], y, col = rgb(0, 0, 1, alpha = 0.5), ...)
# lines(density(x), col = 2, lwd = 2) # Uncomment to add density lines
}
# Menyetarakan berdasarkan formula
pairs(mobilitycovidjakarta$Dirawat~mobilitycovidjakarta$workplaces_percent_change_from_baseline, data = mobilitycovidjakarta,
upper.panel = NULL, # Disabling the upper panel
diag.panel = panel.hist) # Adding the histograms## Warning in par(usr): argument 1 does not name a graphical parameter
## Warning in par(usr): argument 1 does not name a graphical parameter
# plot method
plot(mobilitycovidjakarta$Tanggal, mobilitycovidjakarta$Dirawat, lty = 10, col = 'black', lwd =5)
#plot(mod.ss)
# add lm fit
abline(coef(lm( mobilitycovidjakarta$Dirawat ~ mobilitycovidjakarta$Tanggal , data = mobilitycovidjakarta)), lty = 10, col = 'purple', lwd =5)
rug(mobilitycovidjakarta$Tanggal) # add rug to plot
lines(mobilitycovidjakarta$Tanggal,mod.ss$y , lty = 2, col = 'blue', lwd = 4)
#plot(mod.smsp)
lines(mobilitycovidjakarta$Tanggal, mobilitycovidjakarta$prediksi_model, lty = 2, col = 'cyan', lwd = 4)
legend("topleft",
legend = c("Real", "Model SS", "Trends", "Regresi Nonparametrik"),
lty = 1:4, col = c("Black","blue","purple","cyan"), lwd = 3, bty = "p")
c. Sembuh
# Menambahkan Histograms
panel.hist <- function(x, ...) {
usr <- par("usr")
on.exit(par(usr))
par(usr = c(usr[1:2], 0, 1.5))
his <- hist(x, plot = FALSE)
breaks <- his$breaks
nB <- length(breaks)
y <- his$counts
y <- y/max(y)
rect(breaks[-nB], 0, breaks[-1], y, col = rgb(0, 0, 1, alpha = 0.5), ...)
# lines(density(x), col = 2, lwd = 2) # Uncomment to add density lines
}
# Menyetarakan berdasarkan formula
pairs(mobilitycovidjakarta$Sembuh~mobilitycovidjakarta$workplaces_percent_change_from_baseline, data = mobilitycovidjakarta,
upper.panel = NULL, # Disabling the upper panel
diag.panel = panel.hist) # Adding the histograms## Warning in par(usr): argument 1 does not name a graphical parameter
## Warning in par(usr): argument 1 does not name a graphical parameter
# plot method
plot(mobilitycovidjakarta$Tanggal, mobilitycovidjakarta$Sembuh, lty = 10, col = 'black', lwd =5)
#plot(mod.ss)
# add lm fit
abline(coef(lm( mobilitycovidjakarta$Sembuh ~ mobilitycovidjakarta$Tanggal , data = mobilitycovidjakarta)), lty = 10, col = 'purple', lwd =5)
rug(mobilitycovidjakarta$Tanggal) # add rug to plot
lines(mobilitycovidjakarta$Tanggal,mod.ss$y , lty = 2, col = 'blue', lwd = 4)
#plot(mod.smsp)
lines(mobilitycovidjakarta$Tanggal, mobilitycovidjakarta$prediksi_model, lty = 2, col = 'cyan', lwd = 4)
legend("topleft",
legend = c("Real", "Model SS", "Trends", "Regresi Nonparametrik"),
lty = 1:4, col = c("Black","blue","purple","cyan"), lwd = 3, bty = "p")
d. Meninggal
# Menambahkan Histograms
panel.hist <- function(x, ...) {
usr <- par("usr")
on.exit(par(usr))
par(usr = c(usr[1:2], 0, 1.5))
his <- hist(x, plot = FALSE)
breaks <- his$breaks
nB <- length(breaks)
y <- his$counts
y <- y/max(y)
rect(breaks[-nB], 0, breaks[-1], y, col = rgb(0, 0, 1, alpha = 0.5), ...)
# lines(density(x), col = 2, lwd = 2) # Uncomment to add density lines
}
# Menyetarakan berdasarkan formula
pairs(mobilitycovidjakarta$Meninggal~mobilitycovidjakarta$workplaces_percent_change_from_baseline, data = mobilitycovidjakarta,
upper.panel = NULL, # Disabling the upper panel
diag.panel = panel.hist) # Adding the histograms## Warning in par(usr): argument 1 does not name a graphical parameter
## Warning in par(usr): argument 1 does not name a graphical parameter
# plot method
plot(mobilitycovidjakarta$Tanggal, mobilitycovidjakarta$Meninggal, lty = 10, col = 'black', lwd =5)
#plot(mod.ss)
# add lm fit
abline(coef(lm( mobilitycovidjakarta$Meninggal ~ mobilitycovidjakarta$Tanggal , data = mobilitycovidjakarta)), lty = 10, col = 'purple', lwd =5)
rug(mobilitycovidjakarta$Tanggal) # add rug to plot
lines(mobilitycovidjakarta$Tanggal,mod.ss$y , lty = 2, col = 'blue', lwd = 4)
#plot(mod.smsp)
lines(mobilitycovidjakarta$Tanggal, mobilitycovidjakarta$prediksi_model, lty = 2, col = 'cyan', lwd = 4)
legend("topleft",
legend = c("Real", "Model SS", "Trends", "Regresi Nonparametrik"),
lty = 1:4, col = c("Black","blue","purple","cyan"), lwd = 3, bty = "p")
e. Self Isolation
# Menambahkan Histograms
panel.hist <- function(x, ...) {
usr <- par("usr")
on.exit(par(usr))
par(usr = c(usr[1:2], 0, 1.5))
his <- hist(x, plot = FALSE)
breaks <- his$breaks
nB <- length(breaks)
y <- his$counts
y <- y/max(y)
rect(breaks[-nB], 0, breaks[-1], y, col = rgb(0, 0, 1, alpha = 0.5), ...)
# lines(density(x), col = 2, lwd = 2) # Uncomment to add density lines
}
# Menyetarakan berdasarkan formula
pairs(mobilitycovidjakarta$Self_isolation~mobilitycovidjakarta$workplaces_percent_change_from_baseline, data = mobilitycovidjakarta,
upper.panel = NULL, # Disabling the upper panel
diag.panel = panel.hist) # Adding the histograms## Warning in par(usr): argument 1 does not name a graphical parameter
## Warning in par(usr): argument 1 does not name a graphical parameter
# plot method
plot(mobilitycovidjakarta$Tanggal, mobilitycovidjakarta$Self_isolation, lty = 10, col = 'black', lwd =5)
#plot(mod.ss)
# add lm fit
abline(coef(lm( mobilitycovidjakarta$Self_isolation ~ mobilitycovidjakarta$Tanggal , data = mobilitycovidjakarta)), lty = 10, col = 'purple', lwd =5)
rug(mobilitycovidjakarta$Tanggal) # add rug to plot
lines(mobilitycovidjakarta$Tanggal,mod.ss$y , lty = 2, col = 'blue', lwd = 4)
#plot(mod.smsp)
lines(mobilitycovidjakarta$Tanggal, mobilitycovidjakarta$prediksi_model, lty = 2, col = 'cyan', lwd = 4)
legend("topleft",
legend = c("Real", "Model SS", "Trends", "Regresi Nonparametrik"),
lty = 1:4, col = c("Black","blue","purple","cyan"), lwd = 3, bty = "p")
7. Resrensi
http://users.stat.umn.edu/~helwig/notes/smooth-spline-notes.html
https://www.rumusstatistik.com/2019/06/korelasi-pearson.html
Putu Gita Karlina Jayanti, Rahma Anisa, Muhammad Nur Aidi, Erfiani, Xplore (ISSN:2302-5751), Vol 2 No 2 (2018):15-23, Penerapan Teknik Prapemrosesan Smoothing Spline pada Data Hasil Pengukuran Alat Pemantau Glukosa Darah Non-Invasif