Ejercicio 10 Parcial 3

Se cargan las librerías y los datos

library(kableExtra)
library(dplyr)

## 
## Adjuntando el paquete: 'dplyr'

## The following object is masked from 'package:kableExtra':
## 
##     group_rows

## The following objects are masked from 'package:stats':
## 
##     filter, lag

## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union

library(ggplot2)

load("C:/Users/SANCHEZ/Desktop/GABRIEL2021/Universidad/Ciclo VI/Metodos para el analisis economico/serv_municipales.RData")

dsm <- serv_municipales %>% select(x1,x2,x3,x4,x5,x6)
rownames(dsm) <- serv_municipales$Municipio

## Warning: Setting row names on a tibble is deprecated.

dsm_n <- scale(dsm)

kable(head(dsm_n, 10), align = "c") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed", "responsive"), 
                full_width = F, 
                position = "center") %>%
  row_spec(0, bold = T, background = "#D3D3D3")

	x1	x2	x3	x4	x5	x6
ATIQUIZAYA	2.3520765	1.2410345	2.0750696	1.1964483	1.1964483	1.4300073
EL CARMEN	2.0765626	1.1118237	0.9868413	1.2691683	1.2691683	0.7028941
ALEGRIA	1.8832195	1.1684951	2.2125301	2.0665369	2.0665369	1.6755674
SAN JULIAN	1.8010487	0.4657700	1.2915453	0.6146882	0.6146882	1.5208964
TEJUTLA	1.2430752	1.4031813	1.3069085	1.7183213	1.7183213	0.9702151
PASAQUINA	1.4104851	1.4265679	1.3384228	1.9683992	1.9683992	1.4719561
JUAYUA	1.4861757	1.4133155	1.2044870	1.1964483	1.1964483	2.2480096
SAN SALVADOR	0.0067681	0.3663771	-0.0739910	0.5445109	0.5445109	-0.1551503
SAN PABLO TACACHICO	1.4074575	0.9826130	1.3786035	0.3965282	0.3965282	0.1575592
TEPECOYO	1.4148730	1.2210599	1.1287842	1.5695336	1.5695336	0.0837941

Se determina el número óptimo de grupos utilizando K-means

library(factoextra)

## Welcome! Want to learn more? See two factoextra-related books at https://goo.gl/ve3WBa

fviz_nbclust(dsm_n, kmeans, method = "wss") +
geom_vline(xintercept = 4, linetype = 2)

# Se calcula K_means cluster

set.seed(123)
km_c <- kmeans(dsm_n, centers = 4, nstart = 25)
print(km_c)

## K-means clustering with 4 clusters of sizes 15, 24, 26, 43
## 
## Cluster means:
##           x1         x2        x3         x4         x5         x6
## 1 -1.4346142 -1.6181310 -1.538942 -1.5237906 -1.5237906 -0.9587742
## 2  1.1041991  1.1360869  1.027246  1.2639525  1.2639525  1.0848346
## 3 -0.5709880 -0.4356322 -0.558954 -0.6241510 -0.6241510 -0.6187660
## 4  0.2293982  0.1937748  0.301466  0.2034866  0.2034866  0.1031046
## 
## Clustering vector:
##                ATIQUIZAYA                 EL CARMEN                   ALEGRIA 
##                         2                         2                         2 
##                SAN JULIAN                   TEJUTLA                 PASAQUINA 
##                         2                         2                         2 
##                    JUAYUA              SAN SALVADOR       SAN PABLO TACACHICO 
##                         2                         4                         2 
##                  TEPECOYO         ANTIGUO CUSCATLAN               SAN VICENTE 
##                         2                         2                         4 
##               CIUDAD ARCE                EL PAISNAL         SAN RAFAEL CEDROS 
##                         2                         2                         2 
##             AYUTUXTEPEQUE      SAN FRANCISCO GOTERA         SANTIAGO NONUALCO 
##                         2                         4                         4 
##        SANTA ROSA DE LIMA        SANTA MARIA OSTUMA            CIUDAD BARRIOS 
##                         4                         4                         4 
##                 CHINAMECA          NUEVA CONCEPCION                   JUJUTLA 
##                         2                         2                         4 
##              ZACATECOLUCA                EL ROSARIO                    NEJAPA 
##                         4                         2                         2 
##           NUEVA GUADALUPE                   ARMENIA                  HUIZUCAR 
##                         4                         2                         4 
##        SANTA CRUZ MICHAPA                 SUCHITOTO        SAN PEDRO MASAHUAT 
##                         2                         2                         2 
##         SAN JUAN NONUALCO               APASTEPEQUE                   CORINTO 
##                         4                         4                         3 
##            SAN LUIS TALPA                  ANAMOROS                    BERLIN 
##                         2                         4                         3 
##               EL TRANSITO               LA LIBERTAD               PANCHIMALCO 
##                         3                         4                         4 
##                  MONCAGUA               COJUTEPEQUE       SAN PEDRO PERULAPAN 
##                         4                         4                         2 
##              TEXISTEPEQUE       SAN JOSE VILLANUEVA                CHIRILAGUA 
##                         4                         4                         3 
##                  ILOPANGO             TONACATEPEQUE                  TECOLUCA 
##                         3                         4                         4 
##   SAN BARTOLOME PERULAPIA             SENSUNTEPEQUE               SANTA ELENA 
##                         4                         3                         4 
##                    IZALCO                AHUACHAPAN              CHALATENANGO 
##                         4                         4                         4 
##             SAN SEBASTIAN                SAN MIGUEL                 CONCHAGUA 
##                         4                         4                         4 
##                  EL CONGO                 LOLOTIQUE CANDELARIA DE LA FRONTERA 
##                         1                         4                         4 
##                  LISLIQUE                  USULUTAN                  ZARAGOZA 
##                         4                         3                         3 
##                   METAPAN               SANTA TECLA                 SOYAPANGO 
##                         4                         4                         3 
##                    CALUCO     SAN ANTONIO DEL MONTE                 SONSONATE 
##                         4                         1                         4 
##                 SAN ALEJO                  ILOBASCO      SANTIAGO TEXACUANGOS 
##                         3                         4                         4 
##         SANTIAGO DE MARIA                NAHUIZALCO                     COLON 
##                         3                         3                         3 
##               SANTO TOMAS                COATEPEQUE                 GUAYMANGO 
##                         4                         3                         3 
##                SAN MARTIN                 COMASAGUA                 TAMANIQUE 
##                         1                         3                         3 
##                     APOPA                   DELGADO                   JUCUAPA 
##                         4                         3                         3 
##                  ACAJUTLA                    TACUBA  SAN LUIS DE LA HERRADURA 
##                         3                         3                         1 
##                 OLOCUILTA         PUERTO EL TRIUNFO                JIQUILISCO 
##                         3                         1                         1 
##                SAN MARCOS             QUEZALTEPEQUE                 MEJICANOS 
##                         3                         1                         3 
##                 AGUILARES  SAN SEBASTIAN SALITRILLO                 SANTA ANA 
##                         3                         1                         1 
##             CUSCATANCINGO                   GUAZAPA                CHALCHUAPA 
##                         1                         4                         4 
##                  JUCUARAN                 SONZACATE            SAN JUAN OPICO 
##                         3                         1                         1 
##                  TALNIQUE                  LA UNION    SAN FRANCISCO MENENDEZ 
##                         1                         1                         1 
## 
## Within cluster sum of squares by cluster:
## [1] 24.07670 56.16520 46.38252 72.21429
##  (between_SS / total_SS =  69.0 %)
## 
## Available components:
## 
## [1] "cluster"      "centers"      "totss"        "withinss"     "tot.withinss"
## [6] "betweenss"    "size"         "iter"         "ifault"

aggregate(dsm, by=list(cluster=km_c$cluster), mean)

##   cluster       x1       x2       x3       x4       x5       x6
## 1       1 19.24456 38.51015 35.60929 20.74122 20.74122 14.64705
## 2       2 59.55943 78.47709 72.45514 68.66034 68.66034 42.75283
## 3       3 32.95844 55.66959 49.68016 36.20532 36.20532 19.32319
## 4       4 45.66810 64.80303 62.03424 50.43177 50.43177 29.25109

dsm_c <- cbind(dsm, cluster = km_c$cluster)
head(dsm_c)

##                  x1       x2       x3       x4       x5       x6 cluster
## ATIQUIZAYA 79.37500 80.00000 87.50000 67.50000 67.50000 47.50000       2
## EL CARMEN  75.00000 78.12500 71.87500 68.75000 68.75000 37.50000       2
## ALEGRIA    71.92982 78.94737 89.47368 82.45614 82.45614 50.87719       2
## SAN JULIAN 70.62500 68.75000 76.25000 57.50000 57.50000 48.75000       2
## TEJUTLA    61.76471 82.35294 76.47059 76.47059 76.47059 41.17647       2
## PASAQUINA  64.42308 82.69231 76.92308 80.76923 80.76923 48.07692       2

Se muestra en gráfico

fviz_cluster(km_c, data=dsm_n,
             palette = c("#9a3FDF", "#19AFBB", "#a7B800", "#FC4E07"),
             ellipse.type = "euclid", 
             star.plot = TRUE, 
             repel = TRUE, 
             ggtheme = theme_minimal())