data2

setwd("C:/Users/Jhire/Downloads/final2")

library(rio)
data2 = import("dataOK_all.xlsx")

## New names:
## • `` -> `...1`

str(data2)

## 'data.frame':    196 obs. of  50 variables:
##  $ ...1                   : num  1 2 3 4 5 6 7 8 9 10 ...
##  $ key                    : chr  "AMAZONAS+BAGUA" "AMAZONAS+BONGARA" "AMAZONAS+CHACHAPOYAS" "AMAZONAS+CONDORCANQUI" ...
##  $ Código                 : num  102 103 101 104 105 106 107 202 203 204 ...
##  $ pared1_Ladrillo        : num  4633 1602 3782 291 430 ...
##  $ pared2_Piedra          : num  46 9 22 7 7 7 35 1 0 3 ...
##  $ pared3_Adobe           : num  6639 2729 5881 672 5217 ...
##  $ pared4_Tapia           : num  222 240 2476 8 6052 ...
##  $ pared5_Quincha         : num  2518 157 309 386 346 ...
##  $ pared6_Piedra          : num  127 36 168 7 54 28 518 65 7 6 ...
##  $ pared7_Madera          : num  4484 2505 1270 8145 606 ...
##  $ pared8_Triplay         : num  851 30 91 200 45 24 210 18 0 1 ...
##  $ pared9_Otro            : num  0 0 0 0 0 0 0 0 0 0 ...
##  $ pared10_Total          : num  19520 7308 13999 9716 12757 ...
##  $ techo1_Concreto        : num  2187 692 2262 56 187 ...
##  $ techo2_Madera          : num  294 75 160 188 43 48 340 57 12 8 ...
##  $ techo3_Tejas           : num  179 382 3393 177 3071 ...
##  $ techo4_Planchas        : num  13186 6084 8005 2036 9343 ...
##  $ techo5_Caña            : num  160 38 50 15 26 15 196 10 8 5 ...
##  $ techo6_Triplay         : num  106 5 14 10 12 5 62 17 4 3 ...
##  $ techo7_Paja            : num  3408 32 115 7234 75 ...
##  $ techo8_Otro            : num  0 0 0 0 0 0 0 0 0 0 ...
##  $ techo9_Total           : num  19520 7308 13999 9716 12757 ...
##  $ piso1_Parquet          : num  6 5 23 2 4 3 20 0 0 5 ...
##  $ piso2_Láminas          : num  19 2 36 0 0 4 32 0 0 1 ...
##  $ piso3_Losetas          : num  647 165 1077 20 46 ...
##  $ piso4_Madera           : num  157 132 240 1523 295 ...
##  $ piso5_Cemento          : num  7121 2917 6189 943 1911 ...
##  $ piso6_Tierra           : num  11569 4087 6434 7228 10501 ...
##  $ piso7_Otro             : num  1 0 0 0 0 0 0 0 0 0 ...
##  $ piso8_Total            : num  19520 7308 13999 9716 12757 ...
##  $ agua1_Red              : num  9429 4569 10647 1307 7172 ...
##  $ agua2_Red_fueraVivienda: num  4392 1497 1619 867 3097 ...
##  $ agua3_Pilón            : num  793 215 184 1003 1112 ...
##  $ agua4_Camión           : num  59 0 49 2 0 0 117 0 0 0 ...
##  $ agua5_Pozo             : num  1792 474 876 2564 819 ...
##  $ agua6_Manantial        : num  270 67 92 431 132 211 471 121 61 27 ...
##  $ agua7_Río              : num  2648 388 488 3428 369 ...
##  $ agua8_Otro             : num  56 61 24 80 9 29 104 2 1 6 ...
##  $ agua9_Vecino           : num  81 37 20 34 47 8 177 9 4 6 ...
##  $ agua10_Total           : num  19520 7308 13999 9716 12757 ...
##  $ elec1_Sí               : num  13204 6025 12248 1792 10886 ...
##  $ elec2_No               : num  6316 1283 1751 7924 1871 ...
##  $ elec3_Total            : num  19520 7308 13999 9716 12757 ...
##  $ departamento           : chr  "AMAZONAS" "AMAZONAS" "AMAZONAS" "AMAZONAS" ...
##  $ provincia              : chr  "BAGUA" "BONGARA" "CHACHAPOYAS" "CONDORCANQUI" ...
##  $ Castillo               : num  25629 8374 15671 13154 12606 ...
##  $ Keiko                  : num  10770 5209 10473 1446 7840 ...
##  $ ganaCastillo           : num  1 1 1 1 1 1 1 1 1 1 ...
##  $ countPositivos         : num  8126 389 2174 3481 456 ...
##  $ countFallecidos        : num  462 72 281 111 88 60 336 26 31 21 ...

library(dplyr)

## 
## Attaching package: 'dplyr'

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

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

Utilizando el porcentaje de viviendas que tiene agua de red publica dentro de la vivienda, la razón de votacion de keiko entre castillo, y la tasa fallecidos por cada 1000 contagiados, Ud se propone agrupar a las provincias del Peru (sin la provincia de Lima) siguiendo diversas estrategias (no corrija correlacion negativa si la hubiera, pero siempre normalice); y en ese proceso Ud. encuentra…

Crear la nueva variable

data2$agua_percent <- (data2$agua1_Red / data2$agua10_Total) * 100
data2$tasa_covid <- (data2$countFallecidos / data2$countPositivos) * 1000
data2$razon_presi <- data2$Keiko / data2$Castillo

library(caret)

## Warning: package 'caret' was built under R version 4.3.3

## Loading required package: ggplot2

## Loading required package: lattice

library(factoextra)

## Warning: package 'factoextra' was built under R version 4.3.3

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

columns_to_standardize <- data2[, 51:53]
preProc <- preProcess(columns_to_standardize, method = c("center", "scale"))
standardized_columns <- predict(preProc, columns_to_standardize)
data2[, 51:53] <- standardized_columns
print(head(data2[, 51:53]))

##   agua_percent tasa_covid razon_presi
## 1  -0.35767647 -1.1425581 -0.38149003
## 2   0.38593873 -0.5361040 -0.03106791
## 3   1.09391603 -0.8001609  0.04925502
## 4  -2.18071172 -1.2606334 -0.92027166
## 5   0.05637991 -0.4987786 -0.03127341
## 6   0.20333970  1.1681410  0.08556784

newvar <- subset(data2, provincia != "LIMA")

datanew=newvar[,c(51:53)]
row.names(datanew)=newvar$provincia

library(cluster)

## Warning: package 'cluster' was built under R version 4.3.3

g.dist = daisy(datanew, metric="gower")
library(ggplot2)
library(factoextra)

fviz_nbclust(datanew, pam,diss=g.dist,method = "gap_stat",k.max = 10,verbose = F)

library(kableExtra)

## Warning: package 'kableExtra' was built under R version 4.3.3

## 
## Attaching package: 'kableExtra'

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

set.seed(123)
res.pam=pam(g.dist,5,cluster.only = F)
#nueva columna
datanew$pam=res.pam$cluster
# ver
head(datanew,15)%>%kbl()%>%kable_styling()

	agua_percent	tasa_covid	razon_presi	pam
BAGUA	-0.3576765	-1.1425581	-0.3814900	1
BONGARA	0.3859387	-0.5361040	-0.0310679	1
CHACHAPOYAS	1.0939160	-0.8001609	0.0492550	2
CONDORCANQUI	-2.1807117	-1.2606334	-0.9202717	3
LUYA	0.0563799	-0.4987786	-0.0312734	1
RODRÍGUEZ DE MENDOZA	0.2033397	1.1681410	0.0855678	1
UTCUBAMBA	-0.3364349	-0.9875844	-0.1977399	1
AIJA	1.0259083	0.1450176	-0.0559018	2
ANTONIO RAYMONDI	1.5768436	1.3034891	-0.8405290	2
ASUNCIÓN	0.8467027	0.2718473	-0.6090647	2
BOLOGNESI	0.8979283	0.4646200	-0.2093348	2
CARHUAZ	0.9380403	-0.0149440	-0.3169852	2
CARLOS FERMÍN FITZCARRALD	0.5809679	1.4598788	-0.6551620	2
CASMA	0.2768715	0.3663202	1.2717154	4
CORONGO	1.0471040	1.0170847	0.0549304	2

fviz_silhouette(res.pam,print.summary = F)

silPAM=data.frame(res.pam$silinfo$widths)
silPAM$country=row.names(silPAM)
poorPAM=silPAM[silPAM$sil_width<0,'country']%>%sort()
poorPAM

##  [1] "ACOBAMBA"          "AMBO"              "ANDAHUAYLAS"      
##  [4] "AYMARAES"          "CANGALLO"          "CARAVELÍ"         
##  [7] "CASTROVIRREYNA"    "CAYLLOMA"          "CORONEL PORTILLO" 
## [10] "DATEM DEL MARAÑÓN" "HUAMALÍES"         "HUANCAVELICA"     
## [13] "HUANTA"            "JAUJA"             "OCROS"            
## [16] "OXAPAMPA"          "PUERTO INCA"       "SAN MIGUEL"       
## [19] "SANTA CRUZ"        "TARMA"             "UCAYALI"          
## [22] "VILCAS HUAMÁN"     "YAUYOS"

fviz_nbclust(datanew, hcut,diss=g.dist,method = "gap_stat",k.max = 10,verbose = F,hc_func = "agnes")

set.seed(123)
library(factoextra)
res.agnes<- hcut(g.dist, k = 5,hc_func='agnes',hc_method = "ward.D")
datanew$agnes=res.agnes$cluster
# ver
head(datanew,15)%>%kbl()%>%kable_styling()

	agua_percent	tasa_covid	razon_presi	pam	agnes
BAGUA	-0.3576765	-1.1425581	-0.3814900	1	1
BONGARA	0.3859387	-0.5361040	-0.0310679	1	2
CHACHAPOYAS	1.0939160	-0.8001609	0.0492550	2	2
CONDORCANQUI	-2.1807117	-1.2606334	-0.9202717	3	3
LUYA	0.0563799	-0.4987786	-0.0312734	1	2
RODRÍGUEZ DE MENDOZA	0.2033397	1.1681410	0.0855678	1	2
UTCUBAMBA	-0.3364349	-0.9875844	-0.1977399	1	1
AIJA	1.0259083	0.1450176	-0.0559018	2	2
ANTONIO RAYMONDI	1.5768436	1.3034891	-0.8405290	2	2
ASUNCIÓN	0.8467027	0.2718473	-0.6090647	2	2
BOLOGNESI	0.8979283	0.4646200	-0.2093348	2	2
CARHUAZ	0.9380403	-0.0149440	-0.3169852	2	2
CARLOS FERMÍN FITZCARRALD	0.5809679	1.4598788	-0.6551620	2	2
CASMA	0.2768715	0.3663202	1.2717154	4	4
CORONGO	1.0471040	1.0170847	0.0549304	2	2

fviz_silhouette(res.agnes,print.summary = F)

aaa

silAGNES=data.frame(res.agnes$silinfo$widths)
silAGNES$country=row.names(silAGNES)
poorAGNES=silAGNES[silAGNES$sil_width<0,'country']%>%sort()
poorAGNES

##  [1] "CAJABAMBA"         "CAJATAMBO"         "CARAVELÍ"         
##  [4] "CHINCHEROS"        "CORONEL PORTILLO"  "HUÁNUCO"          
##  [7] "HUAYTARÁ"          "JAÉN"              "JAUJA"            
## [10] "LA UNIÓN"          "LEONCIO PRADO"     "LORETO"           
## [13] "LUYA"              "MANU"              "NAZCA"            
## [16] "OCROS"             "OTUZCO"            "OXAPAMPA"         
## [19] "PACHITEA"          "SÁNCHEZ CARRIÓN"   "SANTIAGO DE CHUCO"
## [22] "SUCRE"             "TARATA"            "TARMA"

fviz_nbclust(datanew, hcut,diss=g.dist,method = "gap_stat",k.max = 10,verbose = F,hc_func = "diana")

set.seed(123)
res.diana <- hcut(g.dist, k = 5,hc_func='diana')
datanew$diana=res.diana$cluster
# veamos
head(datanew,15)%>%kbl%>%kable_styling()

	agua_percent	tasa_covid	razon_presi	pam	agnes	diana
BAGUA	-0.3576765	-1.1425581	-0.3814900	1	1	1
BONGARA	0.3859387	-0.5361040	-0.0310679	1	2	2
CHACHAPOYAS	1.0939160	-0.8001609	0.0492550	2	2	2
CONDORCANQUI	-2.1807117	-1.2606334	-0.9202717	3	3	1
LUYA	0.0563799	-0.4987786	-0.0312734	1	2	2
RODRÍGUEZ DE MENDOZA	0.2033397	1.1681410	0.0855678	1	2	2
UTCUBAMBA	-0.3364349	-0.9875844	-0.1977399	1	1	1
AIJA	1.0259083	0.1450176	-0.0559018	2	2	2
ANTONIO RAYMONDI	1.5768436	1.3034891	-0.8405290	2	2	2
ASUNCIÓN	0.8467027	0.2718473	-0.6090647	2	2	2
BOLOGNESI	0.8979283	0.4646200	-0.2093348	2	2	2
CARHUAZ	0.9380403	-0.0149440	-0.3169852	2	2	2
CARLOS FERMÍN FITZCARRALD	0.5809679	1.4598788	-0.6551620	2	2	2
CASMA	0.2768715	0.3663202	1.2717154	4	4	2
CORONGO	1.0471040	1.0170847	0.0549304	2	2	2

silDIANA=data.frame(res.diana$silinfo$widths)
silDIANA$country=row.names(silDIANA)
poorDIANA=silDIANA[silDIANA$sil_width<0,'country']%>%sort()
poorDIANA

## [1] "ASCOPE"        "CHICLAYO"      "HUACAYBAMBA"   "LA MAR"       
## [5] "LEONCIO PRADO" "PAITA"         "PASCO"         "SECHURA"

fviz_silhouette(res.diana,print.summary = F)

entonces se puede decir que

silDIANA=data.frame(res.diana$silinfo$widths)
silDIANA$country=row.names(silDIANA)
poorDIANA=silDIANA[silDIANA$sil_width<0,'country']%>%sort()
poorDIANA

## [1] "ASCOPE"        "CHICLAYO"      "HUACAYBAMBA"   "LA MAR"       
## [5] "LEONCIO PRADO" "PAITA"         "PASCO"         "SECHURA"