Spatial Garis Kemiskinan Jawa Tengah
Vanessa Indah
October 29, 2024
Library
Input
Input Data
setwd("C:/Unpad/Semester 5/Spatial")
Kemiskinan <- read.csv("Data Spatial Garis Kemiskinan.csv", sep = ";")Input Map
Jateng<-st_read("RBI_50K_2023_Jawa Tengah/RBI_50K_2023_Jawa Tengah.shp")## Reading layer `RBI_50K_2023_Jawa Tengah' from data source
## `C:\Unpad\Semester 5\Spatial\RBI_50K_2023_Jawa Tengah\RBI_50K_2023_Jawa Tengah.shp'
## using driver `ESRI Shapefile'
## Simple feature collection with 35 features and 25 fields
## Geometry type: MULTIPOLYGON
## Dimension: XY, XYZ
## Bounding box: xmin: 108.5558 ymin: -8.211806 xmax: 111.6914 ymax: -5.725371
## z_range: zmin: 0 zmax: 0
## Geodetic CRS: WGS 84 + EGM2008 heightJateng<-st_zm(Jateng)
Jateng<-as_Spatial(Jateng)
plot(Jateng)
Jateng$NAMOBJ## [1] "Banjarnegara" "Banyumas" "Batang" "Blora"
## [5] "Boyolali" "Brebes" "Cilacap" "Demak"
## [9] "Grobogan" "Jepara" "Karanganyar" "Kebumen"
## [13] "Kendal" "Klaten" "Kota Magelang" "Kota Pekalongan"
## [17] "Kota Salatiga" "Kota Semarang" "Kota Surakarta" "Kota Tegal"
## [21] "Kudus" "Magelang" "Pati" "Pekalongan"
## [25] "Pemalang" "Purbalingga" "Purworejo" "Rembang"
## [29] "Semarang" "Sragen" "Sukoharjo" "Tegal"
## [33] "Temanggung" "Wonogiri" "Wonosobo"Data Preparation
Matriks Bobot
id<-c(1:35)
Jateng$id<-c(1:35)
Jateng_sf<-st_as_sf(Jateng)Koordinat Centroid
CoordK<-coordinates(Jateng)Rename
Jateng_merged<-Jateng_sf %>%
left_join(Kemiskinan, by = "NAMOBJ") Mapping Data
ggplot() +
geom_sf(data=Jateng_merged, aes(fill = Kemiskinan),color=NA) +
theme_bw() +
scale_fill_gradient(low = "yellow", high = "red") +
theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank())+
theme(legend.position = "right",
axis.text.x = element_blank(), # Remove x-axis labels
axis.text.y = element_blank())+ # Remove y-axis labels
labs(title = "",
fill = "Garis Kemiskinan") 
Mapping data migrasi masuk di 35 Kabupaten/Kota
di Provinsi Jawa Tengah berdasarkan gradasi warna dari kuning hingga
merah.
Mapping data by Interval
summary(Kemiskinan$Kemiskinan)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 378858 425809 470728 480044 512242 642456breaks <- c(-Inf, 425809, 470728, 512242, Inf)
labels <- c("Very Low", "Low", "High", "Very High")
Jateng_merged$Kemiskinan_Discrete <- cut(Jateng_merged$Kemiskinan, breaks = breaks, labels = labels, right = TRUE)
ggplot() +
geom_sf(data=Jateng_merged, aes(fill = Kemiskinan_Discrete),color=NA) +
theme_bw() +
scale_fill_manual(values = c("Very Low" = "yellow",
"Low" = "orange",
"High" = "red",
"Very High" = "red3"))+
labs(fill = "Garis Kemiskinan")+theme(legend.position = "right",
axis.text.x = element_blank(), # Remove x-axis labels
axis.text.y = element_blank())+ # Remove y-axis labels
labs(title = "",
fill = "Garis Kemiskinan") 
Mapping data migrasi masuk di 35 Kabupaten/Kota
di Provinsi Jawa Tengah berdasarkan kelompok interval (Q1, Q2, dan Q3)
dan membentuk gradasi warna dari kuning hingga merah.
K-Nearest Neighbors
W <- knn2nb(knearneigh(CoordK, k = 5), row.names = id) ; W## Neighbour list object:
## Number of regions: 35
## Number of nonzero links: 175
## Percentage nonzero weights: 14.28571
## Average number of links: 5
## Non-symmetric neighbours listWB <- nb2mat(W, style='B', zero.policy = TRUE) #menyajikan dalam bentuk matrix biner "B"
WB[1:5,1:5]## [,1] [,2] [,3] [,4] [,5]
## 1 0 0 1 0 0
## 2 0 0 0 0 0
## 3 1 0 0 0 0
## 4 0 0 0 0 0
## 5 0 0 0 0 0WL<-nb2listw(W) ; WL # Moran's I (List Neighbours)## Characteristics of weights list object:
## Neighbour list object:
## Number of regions: 35
## Number of nonzero links: 175
## Percentage nonzero weights: 14.28571
## Average number of links: 5
## Non-symmetric neighbours list
##
## Weights style: W
## Weights constants summary:
## n nn S0 S1 S2
## W 35 1225 35 12.44 144.48plot(Jateng, axes=T, col="gray90")
plot(W,coordinates(Jateng),col="red",add=TRUE)
text(CoordK[,1], CoordK[,2], row.names(Jateng), col="black", cex=0.8, pos=1.5)
points(CoordK[,1], CoordK[,2], pch=19, cex=0.7,col="blue") #hubungan antar lokasi
Dengan 35 Kabupaten/Kota di Provinsi Jawa Tengah
yang dianalisis, terdapat hubungan signifikan antar Kabupaten/Kota,
dengan 14.29% hubungan memiliki bobot nonzero dan rata-rata 5 hubungan
per Kabupaten/Kota.
Data Generating
Mengkonversi Koordinat
# Mengkonversi ke format spasial
Jateng <- st_read("RBI_50K_2023_Jawa Tengah/RBI_50K_2023_Jawa Tengah.shp")## Reading layer `RBI_50K_2023_Jawa Tengah' from data source
## `C:\Unpad\Semester 5\Spatial\RBI_50K_2023_Jawa Tengah\RBI_50K_2023_Jawa Tengah.shp'
## using driver `ESRI Shapefile'
## Simple feature collection with 35 features and 25 fields
## Geometry type: MULTIPOLYGON
## Dimension: XY, XYZ
## Bounding box: xmin: 108.5558 ymin: -8.211806 xmax: 111.6914 ymax: -5.725371
## z_range: zmin: 0 zmax: 0
## Geodetic CRS: WGS 84 + EGM2008 heightJateng <- st_zm(Jateng)
# Mengambil koordinat centroid
centroids <- st_centroid(Jateng)
centroid_df <- as.data.frame(st_coordinates(centroids))
centroid_df$NAMOBJ <- Jateng$NAMOBJ
# Menggabungkan data dengan Kemiskinan
Jateng_merged <- Jateng %>%
left_join(Kemiskinan, by = "NAMOBJ")Derajat (Latitude longitude)
data_combined <- merge(Kemiskinan, centroid_df, by = "NAMOBJ")
data_spasial <- data.frame(
y = data_combined[,2],
x1 = data_combined[,3],
x2 = data_combined[,4],
x3 = data_combined[,5],
x4 = data_combined[,6],
longitude = data_combined[,"X"],
latitude = data_combined[,"Y"]
)
coordinates(data_spasial) <- ~longitude+latitudeRegresi Liniear
OLS Model
model_ols <- lm(y ~ x1 + x2 + x3 + x4, data = data_spasial)
summary(model_ols)##
## Call:
## lm(formula = y ~ x1 + x2 + x3 + x4, data = data_spasial)
##
## Residuals:
## Min 1Q Median 3Q Max
## -72239 -39321 -2323 27999 91297
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -487217.4 175664.2 -2.774 0.00944 **
## x1 11473.0 5186.0 2.212 0.03470 *
## x2 10820.9 4832.3 2.239 0.03270 *
## x3 4274.3 22295.5 0.192 0.84926
## x4 488.4 1799.0 0.271 0.78788
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 46140 on 30 degrees of freedom
## Multiple R-squared: 0.5961, Adjusted R-squared: 0.5423
## F-statistic: 11.07 on 4 and 30 DF, p-value: 1.233e-05Dengan taraf signifikansi 5%, diketahui bahwa
variabel tingkat pengangguran terbuka (x1) dan indeks pembangunan
manusia (x2) terbukti signifikan, dengan model secara keseluruhan
menjelaskan 54.23% variasi dalam variabel dependen.
Model Diagnostics
par(mfrow=c(2,2))
plot(model_ols)
Uji Asumsi Klasik
Multikolinearitas
vif(model_ols)## x1 x2 x3 x4
## 1.226246 7.368807 6.617879 1.335688Dengan taraf signifikansi 5%, diketahui bahwa
tidak terdapat multikolinearitas karena 1 < VIF < 10.
Normalitas
resids <- residuals(model_ols)
shapiro.test(resids)##
## Shapiro-Wilk normality test
##
## data: resids
## W = 0.96891, p-value = 0.414Dengan taraf signifikansi 5%, diketahui bahwa
data residual berdistribusi normal karena p-value > 0.05.
Non-Autokorelasi
durbinWatsonTest(model_ols)## lag Autocorrelation D-W Statistic p-value
## 1 -0.1148293 2.193399 0.706
## Alternative hypothesis: rho != 0Dengan taraf signifikansi 5%, diketahui bahwa
tidak terdapat autokorelasi pada residual dari model regresi karena
p-value > 0.05.
Uji Asumsi Efek Spasial
Heterogenitas Spasial
bptest(model_ols)##
## studentized Breusch-Pagan test
##
## data: model_ols
## BP = 10.547, df = 4, p-value = 0.03215Dengan taraf signifikansi 5%, diketahui bahwa
terdapat heteroskedastisitas pada residual dari model regresi karena
p-value < 0.05.
Autokorelasi Spasial
moran_test <- moran.test(resids, WL) ; moran_test##
## Moran I test under randomisation
##
## data: resids
## weights: WL
##
## Moran I statistic standard deviate = 2.3619, p-value = 0.009092
## alternative hypothesis: greater
## sample estimates:
## Moran I statistic Expectation Variance
## 0.189432319 -0.029411765 0.008585353moran_plot <- moran.plot(resids, WL) ; moran_plot
## x wx is_inf labels dfb.1_ dfb.x dffit
## 1 -48111.9429 -5691.7586 FALSE 1 0.034096286 -0.038405211 0.05135676
## 2 -6270.6629 5143.1163 FALSE 2 0.055848972 -0.008198961 0.05644759
## 3 -72239.1623 -662.4676 FALSE 3 0.119212399 -0.201615686 0.23422314
## 4 17947.5455 22490.9158 FALSE 4 0.159156309 0.066874278 0.17263516
## 5 -71505.8920 -13253.0365 TRUE 5 0.008879215 -0.014864372 0.01731445
## 6 67284.6057 9156.0711 FALSE 6 -0.024213327 -0.038141728 -0.04517828
## 7 -69260.1262 27637.7392 FALSE 7 0.378412060 -0.613590231 0.72089435
## 8 21475.4741 25479.4882 FALSE 8 0.178778458 0.089885189 0.20010268
## 9 27204.2809 -4675.5533 FALSE 9 -0.072659682 -0.046276546 -0.08614493
## 10 26286.9372 32063.5450 FALSE 10 0.227649049 0.140099316 0.26730490
## 11 -39817.2573 -21234.9963 FALSE 11 -0.105308057 0.098166460 -0.14396680
## 12 -2645.1966 -20107.2137 FALSE 12 -0.150417125 0.009315055 -0.15070528
## 13 -24638.7913 -4001.2559 FALSE 13 0.011201077 -0.006461143 0.01293099
## 14 -22443.4848 -34861.6076 FALSE 14 -0.244611679 0.128527975 -0.27632284
## 15 34941.0767 -20153.5584 FALSE 15 -0.214140327 -0.175172098 -0.27666106
## 16 50559.0398 -10726.5287 FALSE 16 -0.162905526 -0.192825658 -0.25242810
## 17 -45473.6342 -6321.0547 FALSE 17 0.024741289 -0.026339782 0.03613745
## 18 28793.5095 -12027.7310 FALSE 18 -0.134947317 -0.090968083 -0.16274511
## 19 21313.7946 -45756.6985 TRUE 19 -0.424739322 -0.211940246 -0.47468111
## 20 91296.6535 28903.9917 TRUE 20 0.113071354 0.241678289 0.26682115
## 21 25949.5224 32131.0280 FALSE 21 0.228715522 0.138948927 0.26761464
## 22 -42997.1589 -13013.8077 FALSE 22 -0.033534200 0.033756538 -0.04758200
## 23 56894.9382 31902.7723 FALSE 23 0.184862594 0.246236745 0.30790699
## 24 49544.2509 -6060.4905 FALSE 24 -0.121525590 -0.140958475 -0.18611223
## 25 41813.0019 35455.6088 FALSE 25 0.235721793 0.230749768 0.32986394
## 26 -2323.3892 5035.2272 FALSE 26 0.049047096 -0.002667875 0.04911960
## 27 -43034.7775 -11921.7035 FALSE 27 -0.024597434 0.024782183 -0.03491691
## 28 58597.6468 30856.6448 FALSE 28 0.173520300 0.238045872 0.29457619
## 29 19163.1910 -19248.4198 FALSE 29 -0.178368412 -0.080023209 -0.19549681
## 30 -56191.8094 -20326.0246 FALSE 30 -0.074417378 0.097898845 -0.12297207
## 31 -38825.0492 -24496.2545 FALSE 31 -0.133645528 0.121477721 -0.18060444
## 32 -11798.5111 38360.0418 FALSE 32 0.346561156 -0.095727623 0.35953917
## 33 -30664.8942 -2865.3957 FALSE 33 0.029493546 -0.021173792 0.03630701
## 34 -10028.4330 -27192.7612 FALSE 34 -0.197719526 0.046420822 -0.20309580
## 35 -795.2958 -25973.5394 FALSE 35 -0.201836617 0.003758018 -0.20187160
## cov.r cook.d hat
## 1 1.1344902 1.358354e-03 0.06482060
## 2 1.0882294 1.637528e-03 0.02918720
## 3 1.1629234 2.790157e-02 0.11029319
## 4 1.0438263 1.496641e-02 0.03361575
## 5 1.1929144 1.545674e-04 0.10864256
## 6 1.1795795 1.051823e-03 0.09946776
## 7 0.9123710 2.349782e-01 0.10369200
## 8 1.0321984 1.997296e-02 0.03579377
## 9 1.0957934 3.805335e-03 0.04016099
## 10 0.9957039 3.493992e-02 0.03939255
## 11 1.0981076 1.056576e-02 0.05339904
## 12 1.0440830 1.143603e-02 0.02868100
## 13 1.1052830 8.620654e-05 0.03807816
## 14 0.9766609 3.703484e-02 0.03645954
## 15 1.0172402 3.766751e-02 0.04769042
## 16 1.0824856 3.199073e-02 0.06860183
## 17 1.1310848 6.729396e-04 0.06095402
## 18 1.0684032 1.340100e-02 0.04155462
## 19 0.7784210 9.760916e-02 0.03568543
## 20 1.2354607 3.628252e-02 0.15909894
## 21 0.9944234 3.500344e-02 0.03911654
## 22 1.1257731 1.166047e-03 0.05752298
## 23 1.0794503 4.725830e-02 0.07926345
## 24 1.1062644 1.759493e-02 0.06701103
## 25 1.0075720 5.306097e-02 0.05595027
## 26 1.0892756 1.240895e-03 0.02865596
## 27 1.1270391 6.282532e-04 0.05757366
## 28 1.0919112 4.343093e-02 0.08234299
## 29 1.0308314 1.906599e-02 0.03432222
## 30 1.1406920 7.754045e-03 0.07801825
## 31 1.0815952 1.651288e-02 0.05217710
## 32 0.8633653 5.912584e-02 0.03075138
## 33 1.1095850 6.790781e-04 0.04329711
## 34 1.0109485 2.042159e-02 0.03014635
## 35 1.0058116 2.014104e-02 0.02858133locm <- localmoran(resids,WL) ; locm## Ii E.Ii Var.Ii Z.Ii Pr(z != E(Ii))
## 1 0.150092760 -3.731533e-02 0.2209802223 0.3986682 0.69013768
## 2 -0.017676659 -6.338819e-04 0.0038968626 -0.2730127 0.78484347
## 3 0.026229964 -8.412535e-02 0.4739636122 0.1602953 0.87264848
## 4 0.221244548 -5.192680e-03 0.0317769824 1.2702567 0.20399320
## 5 0.519418723 -8.242617e-02 0.4652519841 0.8823491 0.37758804
## 6 0.337664375 -7.298152e-02 0.4161820787 0.6365401 0.52442443
## 7 -1.049170739 -7.733000e-02 0.4389110496 -1.4669214 0.14239744
## 8 0.299911996 -7.434760e-03 0.0453950124 1.4425290 0.14915319
## 9 -0.069715708 -1.193043e-02 0.0725146217 -0.2145875 0.83008893
## 10 0.461968008 -1.113939e-02 0.0677608132 1.8174833 0.06914314
## 11 0.463429748 -2.555784e-02 0.1532012490 1.2492998 0.21155543
## 12 0.029152125 -1.127970e-04 0.0006937939 1.1110463 0.26654844
## 13 0.054035118 -9.786339e-03 0.0596116676 0.2613973 0.79378615
## 14 0.428842755 -8.120112e-03 0.0495453866 1.9631023 0.04963429
## 15 -0.385965422 -1.968131e-02 0.1186870693 -1.0632034 0.28768970
## 16 -0.297247626 -4.120777e-02 0.2430444451 -0.5193555 0.60351286
## 17 0.157546802 -3.333503e-02 0.1982252023 0.4287313 0.66811881
## 18 -0.189818565 -1.336505e-02 0.0811165102 -0.6195490 0.53555474
## 19 -0.534534616 -7.323236e-03 0.0447190911 -2.4930919 0.01266361
## 20 1.446348864 -1.343666e-01 0.7154961541 1.8687450 0.06165830
## 21 0.456998070 -1.085526e-02 0.0660514173 1.8204077 0.06869695
## 22 0.306693464 -2.980307e-02 0.1778701103 0.7978644 0.42494916
## 23 0.994861056 -5.218296e-02 0.3042533226 1.8982232 0.05766669
## 24 -0.164574074 -3.957018e-02 0.2337845073 -0.2585328 0.79599572
## 25 0.812562307 -2.818410e-02 0.1684884969 2.0482355 0.04053693
## 26 -0.006412117 -8.702126e-05 0.0005352660 -0.2733898 0.78455363
## 27 0.281201903 -2.985524e-02 0.1781719051 0.7369204 0.46117077
## 28 0.991035597 -5.535308e-02 0.3216573063 1.8449998 0.06503757
## 29 -0.202173061 -5.919937e-03 0.0362010007 -1.0314693 0.30232082
## 30 0.626016586 -5.090114e-02 0.2971810026 1.2417248 0.21433811
## 31 0.521281217 -2.429996e-02 0.1458491532 1.4285894 0.15312229
## 32 -0.248065662 -2.244067e-03 0.0137734356 -2.0945893 0.03620751
## 33 0.048159997 -1.515879e-02 0.0918359482 0.2089422 0.83449335
## 34 0.149467497 -1.621241e-03 0.0099569206 1.5141523 0.12998721
## 35 0.011321922 -1.019621e-05 0.0000627215 1.4308789 0.15246492
## attr(,"call")
## localmoran(x = resids, listw = WL)
## attr(,"class")
## [1] "localmoran" "matrix" "array"
## attr(,"quadr")
## mean median pysal
## 1 Low-Low Low-Low Low-Low
## 2 Low-High Low-High Low-High
## 3 Low-High Low-High Low-Low
## 4 High-High High-High High-High
## 5 Low-Low Low-Low Low-Low
## 6 High-High High-High High-High
## 7 Low-High Low-High Low-High
## 8 High-High High-High High-High
## 9 High-Low High-High High-Low
## 10 High-High High-High High-High
## 11 Low-Low Low-Low Low-Low
## 12 Low-Low Low-Low Low-Low
## 13 Low-Low Low-High Low-Low
## 14 Low-Low Low-Low Low-Low
## 15 High-Low High-Low High-Low
## 16 High-Low High-Low High-Low
## 17 Low-Low Low-Low Low-Low
## 18 High-Low High-Low High-Low
## 19 High-Low High-Low High-Low
## 20 High-High High-High High-High
## 21 High-High High-High High-High
## 22 Low-Low Low-Low Low-Low
## 23 High-High High-High High-High
## 24 High-Low High-Low High-Low
## 25 High-High High-High High-High
## 26 Low-High Low-High Low-High
## 27 Low-Low Low-Low Low-Low
## 28 High-High High-High High-High
## 29 High-Low High-Low High-Low
## 30 Low-Low Low-Low Low-Low
## 31 Low-Low Low-Low Low-Low
## 32 Low-High Low-High Low-High
## 33 Low-Low Low-High Low-Low
## 34 Low-Low Low-Low Low-Low
## 35 Low-Low High-Low Low-LowDengan taraf signifikansi 5%, diketahui bahwa
terdapat autokorelasi spasial pada residual dari model regresi karena
p-value < 0.05.
Geographically Weighted Regression (GWR)
Adaptive Kernel Model
bandwidth_optimal <- gwr.sel(
y ~ x1 + x2 + x3 + x4,
data = data_spasial,
coords = cbind(data_spasial$longitude, data_spasial$latitude),
adapt = TRUE)## Adaptive q: 0.381966 CV score: 87338505781
## Adaptive q: 0.618034 CV score: 86084845360
## Adaptive q: 0.763932 CV score: 87105024261
## Adaptive q: 0.5824378 CV score: 85841638787
## Adaptive q: 0.5058644 CV score: 86872922685
## Adaptive q: 0.5813597 CV score: 85848989510
## Adaptive q: 0.5910577 CV score: 85785843627
## Adaptive q: 0.6013617 CV score: 85760260700
## Adaptive q: 0.6020974 CV score: 85774905743
## Adaptive q: 0.5968218 CV score: 85751345850
## Adaptive q: 0.5946201 CV score: 85764266759
## Adaptive q: 0.5982467 CV score: 85743148932
## Adaptive q: 0.5986954 CV score: 85740593823
## Adaptive q: 0.5997138 CV score: 85734841728
## Adaptive q: 0.6003433 CV score: 85740036612
## Adaptive q: 0.5995394 CV score: 85735822504
## Adaptive q: 0.5997895 CV score: 85734417179
## Adaptive q: 0.600001 CV score: 85733257070
## Adaptive q: 0.6001317 CV score: 85735845446
## Adaptive q: 0.5999202 CV score: 85733684072
## Adaptive q: 0.6000509 CV score: 85734245564
## Adaptive q: 0.600001 CV score: 85733257070model_gwr1 <- gwr(
y ~ x1 + x2 + x3 + x4,
data = data_spasial,
coords = cbind(data_spasial$longitude, data_spasial$latitude),
adapt = bandwidth_optimal,
hatmatrix = TRUE,
se.fit = TRUE)
coef_gwr <- model_gwr1$SDF@data[, c("x1", "x2", "x3", "x4")]
summary(model_gwr1)## Length Class Mode
## SDF 35 SpatialPointsDataFrame S4
## lhat 1225 -none- numeric
## lm 11 -none- list
## results 14 -none- list
## bandwidth 35 -none- numeric
## adapt 1 -none- numeric
## hatmatrix 1 -none- logical
## gweight 1 -none- character
## gTSS 1 -none- numeric
## this.call 7 -none- call
## fp.given 1 -none- logical
## timings 12 -none- numericmodel_gwr1## Call:
## gwr(formula = y ~ x1 + x2 + x3 + x4, data = data_spasial, coords = cbind(data_spasial$longitude,
## data_spasial$latitude), adapt = bandwidth_optimal, hatmatrix = TRUE,
## se.fit = TRUE)
## Kernel function: gwr.Gauss
## Adaptive quantile: 0.600001 (about 21 of 35 data points)
## Summary of GWR coefficient estimates at data points:
## Min. 1st Qu. Median 3rd Qu. Max.
## X.Intercept. -642415.427 -596738.749 -510504.220 -379737.827 -293035.809
## x1 8689.713 10185.570 12523.116 13372.485 13994.476
## x2 7547.074 9467.097 10636.669 11720.864 12581.935
## x3 -1124.840 1015.517 9311.947 14487.159 23680.076
## x4 -1534.070 -821.946 -68.455 1285.765 1469.481
## Global
## X.Intercept. -487217.37
## x1 11472.98
## x2 10820.90
## x3 4274.32
## x4 488.41
## Number of data points: 35
## Effective number of parameters (residual: 2traceS - traceS'S): 8.655994
## Effective degrees of freedom (residual: 2traceS - traceS'S): 26.34401
## Sigma (residual: 2traceS - traceS'S): 44870.84
## Effective number of parameters (model: traceS): 7.169543
## Effective degrees of freedom (model: traceS): 27.83046
## Sigma (model: traceS): 43656.1
## Sigma (ML): 38928.8
## AICc (GWR p. 61, eq 2.33; p. 96, eq. 4.21): 861.3293
## AIC (GWR p. 96, eq. 4.22): 846.3595
## Residual sum of squares: 53040811982
## Quasi-global R2: 0.6645492Dengan adaptive quantile 0.600001 dan CV score
85733257070, model GWR efektif dalam menjelaskan 21 titik data (AIC
846.3595 Quasi-global R² 0.6645492).
Fixed Kernel Model
bw1 <- gwr.sel(
y ~ x1 + x2 + x3 + x4,
data = data_spasial,
coords = cbind(data_spasial$longitude,data_spasial$latitude))## Bandwidth: 1.114881 CV score: 88816064524
## Bandwidth: 1.802115 CV score: 88874898868
## Bandwidth: 0.6901475 CV score: 93355239861
## Bandwidth: 1.45408 CV score: 88713224675
## Bandwidth: 1.42018 CV score: 88701837029
## Bandwidth: 1.356895 CV score: 88686932022
## Bandwidth: 1.264454 CV score: 88689065437
## Bandwidth: 1.317623 CV score: 88683504214
## Bandwidth: 1.316234 CV score: 88683486429
## Bandwidth: 1.313435 CV score: 88683474301
## Bandwidth: 1.313762 CV score: 88683474071
## Bandwidth: 1.313803 CV score: 88683474073
## Bandwidth: 1.313722 CV score: 88683474075
## Bandwidth: 1.313762 CV score: 88683474071model_gwr2 <- gwr(
y ~ x1 + x2 + x3 + x4,
data = data_spasial,
coords = cbind(data_spasial$longitude, data_spasial$latitude),
bandwidth = bw1,
hatmatrix = TRUE,
se.fit = TRUE)
coef_gwr2 <- model_gwr2$SDF@data[, c("x1", "x2", "x3", "x4")]
summary(model_gwr2)## Length Class Mode
## SDF 35 SpatialPointsDataFrame S4
## lhat 1225 -none- numeric
## lm 11 -none- list
## results 14 -none- list
## bandwidth 1 -none- numeric
## adapt 0 -none- NULL
## hatmatrix 1 -none- logical
## gweight 1 -none- character
## gTSS 1 -none- numeric
## this.call 7 -none- call
## fp.given 1 -none- logical
## timings 12 -none- numericmodel_gwr2## Call:
## gwr(formula = y ~ x1 + x2 + x3 + x4, data = data_spasial, coords = cbind(data_spasial$longitude,
## data_spasial$latitude), bandwidth = bw1, hatmatrix = TRUE,
## se.fit = TRUE)
## Kernel function: gwr.Gauss
## Fixed bandwidth: 1.313762
## Summary of GWR coefficient estimates at data points:
## Min. 1st Qu. Median 3rd Qu. Max. Global
## X.Intercept. -636210.82 -564750.85 -494367.35 -422970.94 -340589.99 -487217.37
## x1 8823.19 10765.38 12011.65 12910.80 13549.48 11472.98
## x2 9082.05 10044.11 10823.89 11602.38 12209.52 10820.90
## x3 -1184.56 2173.85 6393.37 10140.99 14155.17 4274.32
## x4 -873.30 -392.37 214.17 885.68 1797.39 488.41
## Number of data points: 35
## Effective number of parameters (residual: 2traceS - traceS'S): 7.360732
## Effective degrees of freedom (residual: 2traceS - traceS'S): 27.63927
## Sigma (residual: 2traceS - traceS'S): 45293.43
## Effective number of parameters (model: traceS): 6.334603
## Effective degrees of freedom (model: traceS): 28.6654
## Sigma (model: traceS): 44475.36
## Sigma (ML): 40249.87
## AICc (GWR p. 61, eq 2.33; p. 96, eq. 4.21): 860.7803
## AIC (GWR p. 96, eq. 4.22): 847.8606
## Residual sum of squares: 56701816456
## Quasi-global R2: 0.6413955Dengan fixed bandwidth 1.313762 dan CV score
88683474071, model GWR efektif dalam menjelaskan 35 titik data (AIC
847.8606 Quasi-global R² 0.6413955).
Bi-square
bw2 <- gwr.sel(
y ~ x1 + x2 + x3 + x4,
data = data_spasial,
coords = cbind(data_spasial$longitude,data_spasial$latitude),
gweight = gwr.bisquare)## Bandwidth: 1.114881 CV score: 110797595123
## Bandwidth: 1.802115 CV score: 94429533715
## Bandwidth: 2.226848 CV score: 90824090745
## Bandwidth: 2.322375 CV score: 90402172510
## Bandwidth: 2.548387 CV score: 89801240658
## Bandwidth: 2.67855 CV score: 89647200259
## Bandwidth: 2.756314 CV score: 89584021082
## Bandwidth: 2.816576 CV score: 89534987367
## Bandwidth: 2.85382 CV score: 89505603846
## Bandwidth: 2.876838 CV score: 89487962388
## Bandwidth: 2.891064 CV score: 89477286050
## Bandwidth: 2.899856 CV score: 89470779724
## Bandwidth: 2.90529 CV score: 89466794849
## Bandwidth: 2.908648 CV score: 89464346147
## Bandwidth: 2.910723 CV score: 89462838199
## Bandwidth: 2.912006 CV score: 89461908323
## Bandwidth: 2.912799 CV score: 89461334429
## Bandwidth: 2.913289 CV score: 89460980049
## Bandwidth: 2.913592 CV score: 89460761147
## Bandwidth: 2.913779 CV score: 89460625903
## Bandwidth: 2.913895 CV score: 89460542335
## Bandwidth: 2.913966 CV score: 89460490693
## Bandwidth: 2.91401 CV score: 89460458780
## Bandwidth: 2.91401 CV score: 89460458780model_gwr3 <- gwr(
y ~ x1 + x2 + x3 + x4,
data = data_spasial,
coords = cbind(data_spasial$longitude, data_spasial$latitude),
bandwidth = bw2,
gweight = gwr.bisquare,
hatmatrix = TRUE,
se.fit = TRUE)
coef_gwr3 <- model_gwr3$SDF@data[, c("x1", "x2", "x3", "x4")]
summary(model_gwr3)## Length Class Mode
## SDF 35 SpatialPointsDataFrame S4
## lhat 1225 -none- numeric
## lm 11 -none- list
## results 14 -none- list
## bandwidth 1 -none- numeric
## adapt 0 -none- NULL
## hatmatrix 1 -none- logical
## gweight 1 -none- character
## gTSS 1 -none- numeric
## this.call 8 -none- call
## fp.given 1 -none- logical
## timings 12 -none- numericmodel_gwr3## Call:
## gwr(formula = y ~ x1 + x2 + x3 + x4, data = data_spasial, coords = cbind(data_spasial$longitude,
## data_spasial$latitude), bandwidth = bw2, gweight = gwr.bisquare,
## hatmatrix = TRUE, se.fit = TRUE)
## Kernel function: gwr.bisquare
## Fixed bandwidth: 2.91401
## Summary of GWR coefficient estimates at data points:
## Min. 1st Qu. Median 3rd Qu. Max. Global
## X.Intercept. -664578.52 -562084.32 -493575.57 -424303.72 -324093.64 -487217.37
## x1 8393.64 10641.63 11960.47 12983.60 14240.27 11472.98
## x2 8925.83 10166.15 10841.47 11593.32 12305.73 10820.90
## x3 -1192.41 2422.28 6170.38 9682.00 15624.55 4274.32
## x4 -1151.86 -401.69 226.99 841.74 1915.51 488.41
## Number of data points: 35
## Effective number of parameters (residual: 2traceS - traceS'S): 7.206494
## Effective degrees of freedom (residual: 2traceS - traceS'S): 27.79351
## Sigma (residual: 2traceS - traceS'S): 45438.54
## Effective number of parameters (model: traceS): 6.266229
## Effective degrees of freedom (model: traceS): 28.73377
## Sigma (model: traceS): 44688.91
## Sigma (ML): 40491.33
## AICc (GWR p. 61, eq 2.33; p. 96, eq. 4.21): 860.9707
## AIC (GWR p. 96, eq. 4.22): 848.2109
## Residual sum of squares: 57384175114
## Quasi-global R2: 0.63708Dengan fixed bandwidth 2.91401 dan CV score
89460458780, model GWR efektif dalam menjelaskan 35 titik data (AIC
848.2109 Quasi-global R² 0.63708).
Comparing GWR Model
data.frame("MODEL" = c("Adaptive Gaussian", "Fixed Gaussian", "Bi-Square"),
"AIC" = c(model_gwr1$results$AICh, model_gwr2$results$AICh, model_gwr3$results$AICh),
"R2" = c(0.6645492,0.6413955,0.63708))%>% arrange(AIC)## MODEL AIC R2
## 1 Adaptive Gaussian 846.3595 0.6645492
## 2 Fixed Gaussian 847.8606 0.6413955
## 3 Bi-Square 848.2109 0.6370800Dapat disimpulkan bahwa pemodelan GWR dengan
kernel bandwidth terbaik adalah Fixed Gaussian karena memiliki nilai AIC
terendah dan R2 tertinggi.
GWR Model
# Menggunakan GWR dengan AIC terkecil dan R2 terbesar
model_gwr = model_gwr1
results <-as.data.frame(model_gwr$SDF)
head(results)## sum.w X.Intercept. x1 x2 x3 x4
## 1 21.11540 -642415.4 13994.476 12107.310 772.8846 1430.3832
## 2 23.20282 -609590.0 13331.286 11651.287 1164.2501 1443.5054
## 3 22.81590 -590773.7 13413.683 12300.019 866.7842 769.0229
## 4 21.28866 -340066.3 9044.685 8921.756 15039.6979 -937.1239
## 5 21.91916 -383349.6 10834.337 8933.491 18243.0763 -1073.6904
## 6 23.64115 -595379.3 13104.700 11790.441 346.0111 1325.1237
## X.Intercept._se x1_se x2_se x3_se x4_se gwr.e pred
## 1 183378.5 5502.015 4934.367 22389.06 1900.742 -38984.19 419030.2
## 2 176406.4 5237.872 4816.832 21984.57 1828.188 -5010.27 484037.3
## 3 174109.8 5349.301 4809.355 21894.04 1768.548 -71397.74 450255.7
## 4 180949.9 5583.321 5192.311 23643.19 1889.079 10322.88 414812.1
## 5 180081.8 5881.205 5250.351 23804.19 1885.640 -57043.16 477382.2
## 6 173920.9 5163.243 4781.393 21838.95 1790.301 69638.34 443700.7
## pred.se localR2 X.Intercept._se_EDF x1_se_EDF x2_se_EDF x3_se_EDF
## 1 16693.30 0.6874768 188481.0 5655.109 5071.666 23012.04
## 2 11937.24 0.6566978 181315.0 5383.616 4950.861 22596.29
## 3 14498.34 0.6998012 178954.5 5498.146 4943.176 22503.25
## 4 19790.69 0.7034268 185984.9 5738.678 5336.788 24301.06
## 5 20436.55 0.7468688 185092.7 6044.851 5396.443 24466.55
## 6 23253.65 0.6503113 178760.3 5306.912 4914.436 22446.62
## x4_se_EDF pred.se.1 longitude latitude
## 1 1953.631 17157.80 109.6571 -7.351319
## 2 1879.058 12269.40 109.1758 -7.455506
## 3 1817.758 14901.76 109.8616 -7.020996
## 4 1941.643 20341.37 111.3874 -7.074421
## 5 1938.108 21005.20 110.6525 -7.416504
## 6 1840.116 23900.69 108.9295 -7.062734Mapping the Local R2
gwr.map <- cbind(Jateng, as.matrix(results))
qtm(gwr.map, fill = "localR2") 
Plotting residuals model GWR
gwr.map_sp <- as(gwr.map, "Spatial")
spplot(gwr.map_sp, "localR2")
Comparing OLS and GWR
data.frame("MODEL" = c("GWR","OLS"),
"AIC" = c(model_gwr$results$AICh,AIC(model_ols)),
"R2" = c(0.6645492,0.5423))%>% arrange(AIC)## MODEL AIC R2
## 1 GWR 846.3595 0.6645492
## 2 OLS 857.6854 0.5423000#Attach coefficients to original dataframe
data_spasial$coefx1<-results$x1
data_spasial$coefx2<-results$x2
data_spasial$coefx3<-results$x3
data_spasial$coefx4<-results$x4 Dapat disimpulkan bahwa pemodelan terbaik adalah
GWR Fixed Gaussian karena memiliki nilai AIC terendah dan R² tertinggi,
yaitu 68.3% variasi dalam model dijelaskan oleh variabel
independennya.
Global Significance Test
LMZ.F3GWR.test(model_gwr)##
## Leung et al. (2000) F(3) test
##
## F statistic Numerator d.f. Denominator d.f. Pr(>)
## (Intercept) 3.73018 11.94610 28.698 0.001889 **
## x1 0.68844 15.83113 28.698 0.780481
## x2 0.61915 8.79171 28.698 0.767626
## x3 0.92534 9.19236 28.698 0.519485
## x4 2.38428 7.48818 28.698 0.044373 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Karena p-value untuk variabel x4 < 0.05, maka
terdapat perbedaan signifikan pada x4 antara lokasi satu dengan
lainnya.
Local Significance Test
t1 = model_gwr$SDF$x1 / model_gwr$SDF$x1_se
t2 = model_gwr$SDF$x2 / model_gwr$SDF$x2_se
t3 = model_gwr$SDF$x3 / model_gwr$SDF$x3_se
t4 = model_gwr$SDF$x4 / model_gwr$SDF$x4_se
pvalue1 = 2*pt(q=abs(t1), df=30, lower.tail=FALSE)
pvalue2 = 2*pt(q=abs(t2), df=30, lower.tail=FALSE)
pvalue3 = 2*pt(q=abs(t3), df=30, lower.tail=FALSE)
pvalue4 = 2*pt(q=abs(t4), df=30, lower.tail=FALSE) Variabel x1
df <- data.frame(id = id, t1 = t1, pvalue1 = pvalue1) %>%
mutate(ket = ifelse(pvalue1 < 0.05, "Reject", "Not Reject"))
df## id t1 pvalue1 ket
## 1 1 2.543518 0.01636589 Reject
## 2 2 2.545172 0.01630257 Reject
## 3 3 2.507558 0.01779990 Reject
## 4 4 1.619947 0.11570961 Not Reject
## 5 5 1.842197 0.07535078 Not Reject
## 6 6 2.538075 0.01657588 Reject
## 7 7 2.531973 0.01681423 Reject
## 8 8 1.719448 0.09583610 Not Reject
## 9 9 1.346576 0.18820280 Not Reject
## 10 10 1.881068 0.06970018 Not Reject
## 11 11 1.747473 0.09078583 Not Reject
## 12 12 2.572452 0.01529028 Reject
## 13 13 2.291448 0.02913213 Reject
## 14 14 2.104573 0.04381250 Reject
## 15 15 2.363630 0.02477095 Reject
## 16 16 2.541532 0.01644223 Reject
## 17 17 2.022388 0.05213087 Not Reject
## 18 18 2.021700 0.05220601 Not Reject
## 19 19 1.792552 0.08313481 Not Reject
## 20 20 2.546782 0.01624114 Reject
## 21 21 1.703405 0.09883203 Not Reject
## 22 22 2.334702 0.02644214 Reject
## 23 23 1.722633 0.09525050 Not Reject
## 24 24 2.550486 0.01610066 Reject
## 25 25 2.561516 0.01568893 Reject
## 26 26 2.557777 0.01582738 Reject
## 27 27 2.557109 0.01585226 Reject
## 28 28 1.715003 0.09665847 Not Reject
## 29 29 2.002043 0.05439314 Not Reject
## 30 30 1.520749 0.13879379 Not Reject
## 31 31 1.836160 0.07626276 Not Reject
## 32 32 2.550397 0.01610405 Reject
## 33 33 2.379923 0.02387263 Reject
## 34 34 1.946300 0.06103776 Not Reject
## 35 35 2.552506 0.01602453 RejectVariabel x1 berpengaruh signifikan di
Banjarnegara, Banyumas, Batang, Brebes, Cilacap, Kebumen, Kendal,
Klaten, Kota Magelang, Kota Pekalongan, Kota Tegal, Magelang,
Pekalongan, Pemalang, Purbalingga, Purworejo di Provinsi Jawa
Tengah.
Variabel x2
df2 <- data.frame(id = id, t2 = t2, pvalue2 = pvalue2) %>%
mutate(ket = ifelse(pvalue2 < 0.05, "Reject", "Not Reject"))
df2## id t2 pvalue2 ket
## 1 1 2.453671 0.02016625 Reject
## 2 2 2.418869 0.02184419 Reject
## 3 3 2.557520 0.01583696 Reject
## 4 4 1.718263 0.09605475 Not Reject
## 5 5 1.701503 0.09919225 Not Reject
## 6 6 2.465901 0.01960517 Reject
## 7 7 2.416127 0.02198169 Reject
## 8 8 1.888879 0.06860981 Not Reject
## 9 9 1.318752 0.19723104 Not Reject
## 10 10 2.128294 0.04164126 Reject
## 11 11 1.672271 0.10487031 Not Reject
## 12 12 2.366917 0.02458727 Reject
## 13 13 2.424099 0.02158417 Reject
## 14 14 1.951054 0.06044464 Not Reject
## 15 15 2.197375 0.03585341 Reject
## 16 16 2.586885 0.01477855 Reject
## 17 17 1.934780 0.06249577 Not Reject
## 18 18 2.165001 0.03846977 Reject
## 19 19 1.668690 0.10558413 Not Reject
## 20 20 2.502428 0.01801366 Reject
## 21 21 1.890885 0.06833215 Not Reject
## 22 22 2.149231 0.03980493 Reject
## 23 23 1.905786 0.06630016 Not Reject
## 24 24 2.547085 0.01622962 Reject
## 25 25 2.513092 0.01757196 Reject
## 26 26 2.451401 0.02027198 Reject
## 27 27 2.283678 0.02964071 Reject
## 28 28 1.872949 0.07084925 Not Reject
## 29 29 1.953260 0.06017115 Not Reject
## 30 30 1.431189 0.16271173 Not Reject
## 31 31 1.698996 0.09966904 Not Reject
## 32 32 2.486285 0.01870178 Reject
## 33 33 2.360322 0.02495708 Reject
## 34 34 1.849684 0.07423280 Not Reject
## 35 35 2.416214 0.02197732 RejectVariabel x2 berpengaruh signifikan di
Banjarnegara, Banyumas, Batang, Brebes, Cilacap, Jepara, Kebumen,
Kendal, Kota Magelang, Kota Pekalongan, Kota Semarang, Kota Tegal,
Magelang, Pekalongan, Pemalang, Purbalingga, Purworejo di Provinsi Jawa
Tengah.
Variabel x3
df3 <- data.frame(id = id, t3 = t3, pvalue3 = pvalue3) %>%
mutate(ket = ifelse(pvalue3 < 0.05, "Reject", "Not Reject"))
df3## id t3 pvalue3 ket
## 1 1 0.034520630 0.9726907 Not Reject
## 2 2 0.052957603 0.9581168 Not Reject
## 3 3 0.039589961 0.9686823 Not Reject
## 4 4 0.636111281 0.5295274 Not Reject
## 5 5 0.766380843 0.4494385 Not Reject
## 6 6 0.015843761 0.9874639 Not Reject
## 7 7 0.062659476 0.9504533 Not Reject
## 8 8 0.626273645 0.5358720 Not Reject
## 9 9 0.915407797 0.3672775 Not Reject
## 10 10 0.418783622 0.6783575 Not Reject
## 11 11 0.741920611 0.4639047 Not Reject
## 12 12 0.139439086 0.8900351 Not Reject
## 13 13 0.239605319 0.8122658 Not Reject
## 14 14 0.603826499 0.5504980 Not Reject
## 15 15 0.414131569 0.6817238 Not Reject
## 16 16 -0.050897261 0.9597448 Not Reject
## 17 17 0.628671467 0.5343219 Not Reject
## 18 18 0.467125425 0.6437852 Not Reject
## 19 19 0.767939186 0.4485261 Not Reject
## 20 20 -0.012819955 0.9898563 Not Reject
## 21 21 0.575414080 0.5693025 Not Reject
## 22 22 0.448144884 0.6572685 Not Reject
## 23 23 0.543150422 0.5910395 Not Reject
## 24 24 -0.042096459 0.9667007 Not Reject
## 25 25 -0.026448999 0.9790744 Not Reject
## 26 26 0.022587270 0.9821291 Not Reject
## 27 27 0.285162224 0.7774794 Not Reject
## 28 28 0.533200869 0.5978229 Not Reject
## 29 29 0.617517764 0.5415526 Not Reject
## 30 30 0.873550296 0.3893023 Not Reject
## 31 31 0.748016766 0.4602741 Not Reject
## 32 32 -0.005397868 0.9957289 Not Reject
## 33 33 0.287024860 0.7760665 Not Reject
## 34 34 0.641743032 0.5259135 Not Reject
## 35 35 0.179949755 0.8584020 Not RejectVariabel x3 tidak berpengaruh signifikan di
setiap Kabupaten/Kota di Provinsi Jawa Tengah.
Variabel x4
df4 <- data.frame(id = id, t4 = t4, pvalue4 = pvalue4) %>%
mutate(ket = ifelse(pvalue3 < 0.05, "Reject", "Not Reject"))
df4## id t4 pvalue4 ket
## 1 1 0.752539241 0.4575915 Not Reject
## 2 2 0.789582394 0.4359683 Not Reject
## 3 3 0.434832885 0.6667958 Not Reject
## 4 4 -0.496074399 0.6234570 Not Reject
## 5 5 -0.569403857 0.5733213 Not Reject
## 6 6 0.740168216 0.4649515 Not Reject
## 7 7 0.760063910 0.4531485 Not Reject
## 8 8 -0.530515366 0.5996601 Not Reject
## 9 9 -0.757203736 0.4548343 Not Reject
## 10 10 -0.302799365 0.7641325 Not Reject
## 11 11 -0.546086201 0.5890451 Not Reject
## 12 12 0.704510975 0.4865482 Not Reject
## 13 13 0.002837640 0.9977547 Not Reject
## 14 14 -0.305454433 0.7621295 Not Reject
## 15 15 0.002585021 0.9979546 Not Reject
## 16 16 0.599288833 0.5534796 Not Reject
## 17 17 -0.420826615 0.6768812 Not Reject
## 18 18 -0.347864501 0.7303710 Not Reject
## 19 19 -0.561804204 0.5784229 Not Reject
## 20 20 0.716292981 0.4793497 Not Reject
## 21 21 -0.457401234 0.6506780 Not Reject
## 22 22 -0.038273066 0.9697235 Not Reject
## 23 23 -0.414393191 0.6815343 Not Reject
## 24 24 0.702416654 0.4878342 Not Reject
## 25 25 0.745249936 0.4619198 Not Reject
## 26 26 0.793066386 0.4339668 Not Reject
## 27 27 0.401669587 0.6907744 Not Reject
## 28 28 -0.390438327 0.6989712 Not Reject
## 29 29 -0.432300147 0.6686149 Not Reject
## 30 30 -0.708781497 0.4839319 Not Reject
## 31 31 -0.518569513 0.6078651 Not Reject
## 32 32 0.749073613 0.4596464 Not Reject
## 33 33 0.084182063 0.9334709 Not Reject
## 34 34 -0.375352338 0.7100393 Not Reject
## 35 35 0.434358456 0.6671364 Not RejectVariabel x4 tidak berpengaruh signifikan di
setiap Kabupaten/Kota di Provinsi Jawa Tengah.
Spatial Distribution of Each Variable
map1 <- tm_shape(gwr.map) +
tm_fill("x1",
n = 5,
style = "quantile",
title = "x1 Coefficient") +
tm_borders(col = "black", lwd = 0.5) +
tm_layout(frame = FALSE,
legend.text.size = 0.5,
legend.title.size = 0.6)
map2 <- tm_shape(gwr.map) +
tm_fill("x2",
n = 5,
style = "quantile",
title = "x2 Coefficient") +
tm_borders(col = "black", lwd = 0.5) +
tm_layout(frame = FALSE,
legend.text.size = 0.5,
legend.title.size = 0.6)
map3 <- tm_shape(gwr.map) +
tm_fill("x3",
n = 5,
style = "quantile",
title = "x3 Coefficient") +
tm_borders(col = "black", lwd = 0.5) +
tm_layout(frame = FALSE,
legend.text.size = 0.5,
legend.title.size = 0.6)
map4 <- tm_shape(gwr.map) +
tm_fill("x4",
n = 5,
style = "quantile",
title = "x4 Coefficient") +
tm_borders(col = "black", lwd = 0.5) +
tm_layout(frame = FALSE,
legend.text.size = 0.5,
legend.title.size = 0.6)
tmap_arrange(map1, map2, map3, map4, ncol = 2, nrow = 2)
Membuat plot -spplot dan -ggplot2
Ekstrak koefisien lokal untuk variabel x1, x2, x3
# Ekstrak koefisien lokal untuk variabel x1, x2, x3, x4
coef_x1 <- model_gwr$SDF$x1
coef_x2 <- model_gwr$SDF$x2
coef_x3 <- model_gwr$SDF$x3
coef_x4 <- model_gwr$SDF$x4
# Konversi SpatialPointsDataFrame menjadi data frame biasa
gwr_df <- as.data.frame(model_gwr$SDF)Plot koefisien regresi lokal untuk variabel x1
# Koefisien Lokal x1 dengan ssplot
spplot(model_gwr$SDF, "x1", main = "Koefisien GWR untuk x1 (ssplot)", col.regions = terrain.colors(100))
# Koefisien Lokal x1 dengan ggplot2
ggplot(gwr_df, aes(x = data_spasial$longitude, y = data_spasial$latitude)) +
geom_point(aes(color = x1), size = 2) +
scale_color_viridis_c() +
labs(title = "Koefisien GWR untuk x1 (ggplot2)", color = "Koefisien") +
theme_minimal()
Plot koefisien regresi lokal untuk variabel x2
# Koefisien Lokal x2 dengan ssplot
spplot(model_gwr$SDF, "x2", main = "Koefisien GWR untuk x2 (ssplot)", col.regions = terrain.colors(100))
# Koefisien Lokal x2 dengan ggplot2
ggplot(gwr_df, aes(x = data_spasial$longitude, y = data_spasial$latitude)) +
geom_point(aes(color = x2), size = 2) +
scale_color_viridis_c() +
labs(title = "Koefisien GWR untuk x2 (ggplot2)", color = "Koefisien") +
theme_minimal()
Plot koefisien regresi lokal untuk variabel x3
# Koefisien Lokal x3 dengan ssplot
spplot(model_gwr$SDF, "x3", main = "Koefisien GWR untuk x3 (ssplot)", col.regions = terrain.colors(100))
# Koefisien Lokal x3 dengan ggplot2
ggplot(gwr_df, aes(x = data_spasial$longitude, y = data_spasial$latitude)) +
geom_point(aes(color = x3), size = 2) +
scale_color_viridis_c() +
labs(title = "Koefisien GWR untuk x3 (ggplot)", color = "Koefisien") +
theme_minimal()
Plot koefisien regresi lokal untuk variabel x4
# Koefisien Lokal x4 dengan ssplot
spplot(model_gwr$SDF, "x4", main = "Koefisien GWR untuk x4 (ssplot)", col.regions = terrain.colors(100))
# Koefisien Lokal x4 dengan ggplot2
ggplot(gwr_df, aes(x = data_spasial$longitude, y = data_spasial$latitude)) +
geom_point(aes(color = x4), size = 2) +
scale_color_viridis_c() +
labs(title = "Koefisien GWR untuk x3 (ggplot)", color = "Koefisien") +
theme_minimal()