Regresi Linear Berganda Penyebaran KASUS POSITIF Covid 19 DKI Jakarta Bulan September 2021
Author : Cak Isa
Tanggal : 2022-03-28
Dosen Pengampu : Prof. Dr. Suhartono, Mkom
Magister Informatika - UIN Maulana Malik Ibrahim Malang
Kasus POSITIF COVID 19
Melakukan Analisa menggunakan Regresi Linear Berganda Covid 19 di DKI Jakarta pada bulan September 2021 dengan memandingkan data dari Google Mobility Index dan Data Riwayat/Penyebaran Covid 19 DKI Jakarta untuk kasus POSITIF Covid 19
Sumber Data :
1. Mobility Index https://www.google.com/covid19/mobility/?hl=id
2. Riwayat Covid 19 DKI Jakarta https://riwayat-file-covid-19-dki-jakarta-jakartagis.hub.arcgis.com/
library(readxl)
library(reshape2)## Warning: package 'reshape2' was built under R version 4.1.3#baca data excel
mobindexjkt <- read_excel("mobindex_jkt.xlsx")
covidjkt <- read_excel("c19data_jkt.xlsx")
retail <- mobindexjkt$retail_and_recreation_percent_change_from_baseline
grocery <- mobindexjkt$grocery_and_pharmacy_percent_change_from_baseline
park <- mobindexjkt$parks_percent_change_from_baseline
station <- mobindexjkt$transit_stations_percent_change_from_baseline
workplace <- mobindexjkt$workplaces_percent_change_from_baseline
residental <- mobindexjkt$residential_percent_change_from_baseline
Tgl <- covidjkt$Tanggal
Positif <- covidjkt$Positif
Dirawat <- covidjkt$Dirawat
Sembuh <- covidjkt$Sembuh
Meninggal <- covidjkt$Meninggal
Isolasi_Mandiri <- covidjkt$SelfIsolation
dataku1 <- data.frame(Positif, retail, grocery, park, station, workplace, residental)
dataku2 <- melt(data = dataku1, id.vars = "Positif")Inisialisasi Data per Kolom
#dataframe KASUS POSITIF
library(ggplot2)
library(reshape2)
library(tidyr)##
## Attaching package: 'tidyr'## The following object is masked from 'package:reshape2':
##
## smithsdataku3 <- data.frame(Tgl, retail, grocery, park, station, workplace, residental)
dataku4 <- melt(data = dataku3, id.vars = "Tgl")
ggplot(data = dataku4, aes(x = Tgl, y = value, colour = variable))+
geom_point() +
geom_line() +
theme(legend.justification = "top") +
labs(title = "Google Mobility Index Covid 19",
subtitle = "Propinsi DKI Jakarta Indonesia SEPT 2021",
y = "Mobility Index", x = "Tanggal") +
theme(axis.text.x = element_text(angle = -90))
Dari Gambar diatas mobiity index terbanyak pada RETAIL walaupun
cenderung turun, dikarenakan banyak mall yang menerapkan aturan ketat
atau bahkan ada yang tutup
dataku5 <- data.frame(Tgl, Positif)
dataku6 <- melt(data = dataku5, id.vars = "Tgl")
ggplot(data = dataku6, aes(x = Tgl, y = value, colour = variable))+
geom_point() +
geom_line() +
theme(legend.justification = "top") +
labs(title = "Jumlah Kasus Positif Covid 19",
subtitle = "Propinsi DKI Jakarta Indonesia SEPT 2021",
y = "Jumlah Kasus", x = "Tanggal") +
theme(axis.text.x = element_text(angle = -90))
Kasus Positif terus berkembang dari hari ke hari
library(gridExtra)## Warning: package 'gridExtra' was built under R version 4.1.3p1=ggplot(dataku1, aes(x = Positif, y = retail)) +
geom_point() +
stat_smooth()
p2=ggplot(data = dataku1, aes(x = Positif, y = grocery)) +
geom_point() +
stat_smooth()
p3=ggplot(data = dataku1, aes(x = Positif, y = park)) +
geom_point() +
stat_smooth()
p4=ggplot(data = dataku1, aes(x = Positif, y = station)) +
geom_point() +
stat_smooth()
p5=ggplot(data = dataku1, aes(x = Positif, y = workplace)) +
geom_point() +
stat_smooth()
p6=ggplot(data = dataku1, aes(x = Positif, y = residental)) +
geom_point() +
stat_smooth()
gridExtra::grid.arrange(p1,p2,p3,p4, p5, p6, ncol=2)## `geom_smooth()` using method = 'loess' and formula 'y ~ x'
## `geom_smooth()` using method = 'loess' and formula 'y ~ x'
## `geom_smooth()` using method = 'loess' and formula 'y ~ x'
## `geom_smooth()` using method = 'loess' and formula 'y ~ x'
## `geom_smooth()` using method = 'loess' and formula 'y ~ x'
## `geom_smooth()` using method = 'loess' and formula 'y ~ x'
model <- lm(Positif ~ retail+grocery+park+station+workplace+residental)
summary(model)##
## Call:
## lm(formula = Positif ~ retail + grocery + park + station + workplace +
## residental)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1552.42 -399.60 -33.81 646.02 1192.06
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 879114.81 3089.35 284.563 < 2e-16 ***
## retail 771.83 156.99 4.917 5.74e-05 ***
## grocery -466.08 74.17 -6.284 2.06e-06 ***
## park -210.43 85.22 -2.469 0.02139 *
## station 426.70 138.44 3.082 0.00526 **
## workplace -64.36 71.57 -0.899 0.37781
## residental -136.46 289.40 -0.472 0.64169
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 815.3 on 23 degrees of freedom
## Multiple R-squared: 0.8597, Adjusted R-squared: 0.8231
## F-statistic: 23.49 on 6 and 23 DF, p-value: 9.935e-09Digaram PLOT menunjukkan terjadinya angka positif bebentuk curva hampir 45 derajad pada retail, park dan station. Dari Analisa Linear Model R Square 80% Adjusted Squared 80% dan p value diatas 5 %, yang artinya tingkat errornya rendah. Dan untuk park/taman mempunyai nilai yang paling significant yaitu 0.02139
residu <- residuals(model)
qqnorm(residu)
Pada Diagram QQPLOT berbentuk curva naik 45 derajat artinya tingkat
error untuk analisa ini rendah
cor(dataku2$value, dataku2$Positif)## [1] 0.0314721anova(model)## Analysis of Variance Table
##
## Response: Positif
## Df Sum Sq Mean Sq F value Pr(>F)
## retail 1 13699582 13699582 20.6119 0.0001465 ***
## grocery 1 41074068 41074068 61.7986 5.784e-08 ***
## park 1 307605 307605 0.4628 0.5031038
## station 1 37929330 37929330 57.0671 1.131e-07 ***
## workplace 1 517677 517677 0.7789 0.3866157
## residental 1 147787 147787 0.2224 0.6416942
## Residuals 23 15286813 664644
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1plot(dataku2$Positif ~ dataku2$value, data = dataku1, col = "dodgerblue", pch = 20, cex = 1.5, main = "Mobility Index & Kasus Positif Covid 19 DKI JKT Sept 21")
abline(model) #Add a regression line## Warning in abline(model): only using the first two of 7 regression coefficients
plot(cooks.distance(model), pch = 16, col = "blue") #Plot the Cooks Distances.
plot(model)
AIC(model)## [1] 495.3754BIC(model)## [1] 506.585head(predict(model), n = 30)## 1 2 3 4 5 6 7 8
## 851702.8 852093.2 852069.7 852387.8 852739.9 852935.9 854147.1 855151.4
## 9 10 11 12 13 14 15 16
## 854422.8 854488.8 854785.5 854155.3 855818.8 854429.6 855800.8 854953.3
## 17 18 19 20 21 22 23 24
## 856007.3 855352.8 855004.3 855059.9 856195.5 856834.9 856470.0 856262.3
## 25 26 27 28 29 30
## 858119.2 858559.6 856691.1 856499.9 857349.6 856866.7plot(head(predict(model), n = 30))
head(resid(model), n = 30)## 1 2 3 4 5 6
## -446.843832 -407.194850 -40.721194 1.215545 -47.872286 -26.897950
## 7 8 9 10 11 12
## -976.148348 -1552.416916 -515.808175 -320.798402 -349.509578 586.654798
## 13 14 15 16 17 18
## -889.837939 689.390895 -376.811364 646.746701 -201.347333 653.239847
## 19 20 21 22 23 24
## 1156.747600 1192.060377 162.542453 -249.928985 279.987424 668.680647
## 25 26 27 28 29 30
## -1055.239876 -1327.567515 643.854243 939.149486 266.370710 898.303818coef(model)## (Intercept) retail grocery park station workplace
## 879114.81268 771.82559 -466.08349 -210.42662 426.70483 -64.36163
## residental
## -136.46322dataku1$residuals <- model$residuals
dataku1$predicted <- model$fitted.values
dataku1## Positif retail grocery park station workplace residental residuals
## 1 851256 -21 8 -45 -41 -34 12 -446.843832
## 2 851686 -20 7 -43 -42 -34 12 -407.194850
## 3 852029 -18 8 -37 -41 -32 13 -40.721194
## 4 852389 -21 9 -39 -35 -19 6 1.215545
## 5 852692 -24 3 -46 -38 -14 6 -47.872286
## 6 852909 -21 4 -42 -41 -34 12 -26.897950
## 7 853171 -24 -1 -46 -40 -35 13 -976.148348
## 8 853599 -24 -3 -46 -40 -34 12 -1552.416916
## 9 853907 -20 4 -41 -39 -33 11 -515.808175
## 10 854168 -17 6 -36 -39 -31 12 -320.798402
## 11 854436 -21 6 -40 -33 -18 6 -349.509578
## 12 854742 -25 0 -46 -36 -13 6 586.654798
## 13 854929 -21 1 -45 -39 -34 12 -889.837939
## 14 855119 -28 -8 -55 -44 -38 15 689.390895
## 15 855424 -20 3 -43 -38 -34 11 -376.811364
## 16 855600 -20 3 -41 -39 -34 11 646.746701
## 17 855806 -16 5 -35 -38 -32 12 -201.347333
## 18 856006 -20 6 -37 -32 -18 6 653.239847
## 19 856161 -28 -5 -50 -36 -15 7 1156.747600
## 20 856252 -20 4 -42 -38 -33 11 1192.060377
## 21 856358 -22 -1 -44 -38 -34 12 162.542453
## 22 856585 -20 -1 -42 -39 -31 11 -249.928985
## 23 856750 -19 1 -39 -38 -31 11 279.987424
## 24 856931 -15 6 -33 -37 -29 11 668.680647
## 25 857064 -16 9 -36 -29 -16 5 -1055.239876
## 26 857232 -20 1 -43 -32 -12 6 -1327.567515
## 27 857335 -18 4 -41 -37 -31 11 643.854243
## 28 857439 -18 5 -42 -37 -32 11 939.149486
## 29 857616 -17 5 -40 -36 -31 10 266.370710
## 30 857765 -15 10 -38 -34 -29 10 898.303818
## predicted
## 1 851702.8
## 2 852093.2
## 3 852069.7
## 4 852387.8
## 5 852739.9
## 6 852935.9
## 7 854147.1
## 8 855151.4
## 9 854422.8
## 10 854488.8
## 11 854785.5
## 12 854155.3
## 13 855818.8
## 14 854429.6
## 15 855800.8
## 16 854953.3
## 17 856007.3
## 18 855352.8
## 19 855004.3
## 20 855059.9
## 21 856195.5
## 22 856834.9
## 23 856470.0
## 24 856262.3
## 25 858119.2
## 26 858559.6
## 27 856691.1
## 28 856499.9
## 29 857349.6
## 30 856866.7scatter.smooth(x=dataku2$value, y=dataku2$Positif, main="Mobiliy Index ~ Positif")
boxplot(dataku2$value, main="Covid DKI JKT Sept 21", boxplot.stats(dataku2$value)$out)
plot(density(dataku2$value), main="Data Covid DKI JKT Sept 21", ylab="Frequency")