library(readxl)
Democracia_IDH_AL <- read_excel("Democracia_IDH_AL.xlsx")
View(Democracia_IDH_AL)#Observado el tipo de variables de la data del IDH y democracia en AL
str(Democracia_IDH_AL)## tibble [20 × 8] (S3: tbl_df/tbl/data.frame)
## $ País : chr [1:20] "Argentina" "Bolivia" "Brasil" "Chile" ...
## $ Puntuación : num [1:20] 7.02 5.7 6.97 7.97 6.96 8.07 3 6.27 5.96 5.6 ...
## $ Procesoelectoralypluralismo: num [1:20] 9.17 7.5 9.58 9.58 9.17 9.58 1.08 8.75 9.17 7.92 ...
## $ Funcionamientodelgobierno : num [1:20] 5.36 4.64 5.36 8.57 6.79 7.5 3.57 5.36 4.29 5.36 ...
## $ Participaciónpolítica : num [1:20] 6.11 5.56 6.67 4.44 5 6.67 3.33 6.11 5.56 3.89 ...
## $ Culturapolítica : num [1:20] 6.25 3.75 5 8.13 5.63 7.5 4.38 4.38 3.75 4.38 ...
## $ Derechosciviles : num [1:20] 8.24 7.06 8.24 9.12 8.24 9.12 2.65 7.65 7.06 6.47 ...
## $ IDH : chr [1:20] "0,841" "0,718" "0,758" "0,849" ...#índice de democracia en AL
str(Democracia_IDH_AL$IDH)## chr [1:20] "0,841" "0,718" "0,758" "0,849" "0,756" "0,811" "0,759" "0,734" ...library(DescTools)
library(moments)Democracia_IDH_AL$IDH <- as.numeric(gsub(",", ".", Democracia_IDH_AL$IDH))
print(Democracia_IDH_AL)## # A tibble: 20 × 8
## País Puntuación Procesoelectoralypluralismo Funcionamientodelgobie…¹
## <chr> <dbl> <dbl> <dbl>
## 1 Argentina 7.02 9.17 5.36
## 2 Bolivia 5.7 7.5 4.64
## 3 Brasil 6.97 9.58 5.36
## 4 Chile 7.97 9.58 8.57
## 5 Colombia 6.96 9.17 6.79
## 6 Costa Rica 8.07 9.58 7.5
## 7 Cuba 3 1.08 3.57
## 8 Ecuador 6.27 8.75 5.36
## 9 El Salvador 5.96 9.17 4.29
## 10 Guatemala 5.6 7.92 5.36
## 11 Haití 4.91 5.58 2.93
## 12 Honduras 5.63 8.5 4.64
## 13 México 6.19 8.33 6.07
## 14 Nicaragua 3.63 2.67 1.86
## 15 Panamá 7.05 9.58 6.07
## 16 Paraguay 6.24 8.75 5.71
## 17 Perú 6.6 9.17 5
## 18 R. Dominicana 6.54 9.17 5.36
## 19 Uruguay 8.38 10 8.57
## 20 Venezuela 3.16 1.67 1.79
## # ℹ abbreviated name: ¹Funcionamientodelgobierno
## # ℹ 4 more variables: Participaciónpolítica <dbl>, Culturapolítica <dbl>,
## # Derechosciviles <dbl>, IDH <dbl>str(Democracia_IDH_AL$IDH)## num [1:20] 0.841 0.718 0.758 0.849 0.756 0.811 0.759 0.734 0.676 0.645 ...head(Democracia_IDH_AL$IDH)## [1] 0.841 0.718 0.758 0.849 0.756 0.811Democracia_IDH_AL$IDH <- gsub("[^0-9.]", "", Democracia_IDH_AL$IDH)
Democracia_IDH_AL$IDH <- as.numeric(Democracia_IDH_AL$IDH)
summary(Democracia_IDH_AL$IDH)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.5590 0.6873 0.7570 0.7381 0.7880 0.8490allStats=c(summary(Democracia_IDH_AL$IDH),
sd=sd(Democracia_IDH_AL$IDH),
skew=Skew(Democracia_IDH_AL$IDH),
kurt=Kurt(Democracia_IDH_AL$IDH),
cv=CoefVar(Democracia_IDH_AL$IDH))
allStats## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.55900000 0.68725000 0.75700000 0.73810000 0.78800000 0.84900000
## sd skew kurt cv
## 0.07579467 -0.54948487 -0.47384885 0.10268889Resultado: se observan los puntajes del índice de desarrollo humano, el mayor puntajes es 0.849 (Chile), mientras que el mínimo es 0.559(Haití)
library(ggplot2)
base=ggplot(data=Democracia_IDH_AL,
aes(x=IDH))
histogram= base + geom_histogram(aes(y = after_stat(density)),
colour = 1, fill = "skyblue",bins=10) +
stat_function(fun = dnorm,
args = list(mean = allStats['Mean'],
sd = allStats['sd']),col='pink')
histogram
RESULTADO: Se observa que la densidad del puntaje del IDH se encuentra
entre 700 y 800 de puntaje, el promedio de todo el puntaje del IDH en
América Latina es 738.1
#Análisis bivariado
str(Democracia_IDH_AL$Puntuación)## num [1:20] 7.02 5.7 6.97 7.97 6.96 8.07 3 6.27 5.96 5.6 ...#Análisis de correlación entre el IDH la Puntuación de democracia y participación política
library(ggplot2)
library(ggrepel)names(Democracia_IDH_AL) #verifico los nombres## [1] "País" "Puntuación"
## [3] "Procesoelectoralypluralismo" "Funcionamientodelgobierno"
## [5] "Participaciónpolítica" "Culturapolítica"
## [7] "Derechosciviles" "IDH"ggplot(Democracia_IDH_AL, aes(x = Puntuación, y = IDH, label = País)) +
geom_point() +
geom_text_repel(size = 3) +
labs(x = "Puntuación", y = "IDH", title = "Gráfico de dispersión democracia e IDH en AL")
#Índice de correlación con matriz
f1 <- Democracia_IDH_AL[, c("IDH", "Puntuación", "Participaciónpolítica", "Culturapolítica")]correlation_matrix <- cor(f1)
correlation_matrix## IDH Puntuación Participaciónpolítica
## IDH 1.0000000 0.6249085 0.57783965
## Puntuación 0.6249085 1.0000000 0.66445396
## Participaciónpolítica 0.5778396 0.6644540 1.00000000
## Culturapolítica 0.4007599 0.5238965 -0.01544892
## Culturapolítica
## IDH 0.40075990
## Puntuación 0.52389652
## Participaciónpolítica -0.01544892
## Culturapolítica 1.00000000RESULTADO: En cuanto al IDH y la Puntuación de democracia se observa que hay una correlación positiva. Puede indicar que los países con puntuaciones de democracia más altos influyen en el IDH y por ende esta variable es más alta. Asimismo, el cuando al IDH y la Participación Política, se observa que también hay una correlación positiva moderada. Por ende, se puede proponer que los países con un mayor puntaje en participación política tienen un IDH más alto. Finalmente, se observa que en cuanto al IDH y la Cultura política, hay una baja correlación entre ambas variables; sin embargo, aporta a la correlación.
#REGRESIÓN #modelo1
f1modelo <- formula(IDH ~ Puntuación + Participaciónpolítica)
modelo1 <- lm(f1, data = Democracia_IDH_AL)
summary(modelo1)##
## Call:
## lm(formula = f1, data = Democracia_IDH_AL)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.14427 -0.02483 0.00174 0.02006 0.12191
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.416513 0.088304 4.717 0.000233 ***
## Puntuación 0.004084 0.017033 0.240 0.813570
## Participaciónpolítica 0.035460 0.019218 1.845 0.083606 .
## Culturapolítica 0.020603 0.014158 1.455 0.164944
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.0582 on 16 degrees of freedom
## Multiple R-squared: 0.5036, Adjusted R-squared: 0.4105
## F-statistic: 5.41 on 3 and 16 DF, p-value: 0.009197#modelo2
f1modelo2 <- formula(IDH ~ Puntuación + Participaciónpolítica + Culturapolítica) #incluyendo otra variable
modelo2 <- lm(f1, data = Democracia_IDH_AL)
summary(modelo2)##
## Call:
## lm(formula = f1, data = Democracia_IDH_AL)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.14427 -0.02483 0.00174 0.02006 0.12191
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.416513 0.088304 4.717 0.000233 ***
## Puntuación 0.004084 0.017033 0.240 0.813570
## Participaciónpolítica 0.035460 0.019218 1.845 0.083606 .
## Culturapolítica 0.020603 0.014158 1.455 0.164944
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.0582 on 16 degrees of freedom
## Multiple R-squared: 0.5036, Adjusted R-squared: 0.4105
## F-statistic: 5.41 on 3 and 16 DF, p-value: 0.009197RESULTADO:
#Haciendo un modelo alternativo con el uso de otras variables
f1modeloalt <- formula(IDH ~ Puntuación + Participaciónpolítica + Culturapolítica + Derechosciviles + Funcionamientodelgobierno)
modeloalt <- lm(f1modeloalt, data = Democracia_IDH_AL)
summary(modeloalt)##
## Call:
## lm(formula = f1modeloalt, data = Democracia_IDH_AL)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.085684 -0.029336 0.003183 0.025101 0.086983
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.437040 0.070214 6.224 2.22e-05 ***
## Puntuación -0.081243 0.070813 -1.147 0.27049
## Participaciónpolítica 0.058024 0.018815 3.084 0.00809 **
## Culturapolítica 0.027010 0.011593 2.330 0.03530 *
## Derechosciviles 0.009498 0.039006 0.243 0.81116
## Funcionamientodelgobierno 0.053304 0.019031 2.801 0.01415 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.04592 on 14 degrees of freedom
## Multiple R-squared: 0.7295, Adjusted R-squared: 0.6329
## F-statistic: 7.551 on 5 and 14 DF, p-value: 0.001259RESULTADO: Dado que la variable Puntuación, Participaciónpolítica y Culturapolítica no aportan en los primeros dos modelos de regresión, se crea un tercer modelo alternativo que conecta a todas las variables independientes con la variable dependiente. De ese modo se observa que hay significancia de la variable Participaciónpolítica, Culturapolítica y Funcionamientodelgobierno en el Índice de desarrollo humano a nivel de América Latina ya que, son menores a 0.05. Además, el efecto es directo dado que en la columna Estimate las variables son positivas.
#Vemos la gráfica de correlación entre las variables
library(ggcorrplot)
colNums=names(Democracia_IDH_AL)[c(4,5,6,8)]
numXs=Democracia_IDH_AL[,colNums]
ggcorrplot(cor(numXs),lab = T,show.diag = F)
library(modelsummary)## `modelsummary` 2.0.0 now uses `tinytable` as its default table-drawing
## backend. Learn more at: https://vincentarelbundock.github.io/tinytable/
##
## Revert to `kableExtra` for one session:
##
## options(modelsummary_factory_default = 'kableExtra')
## options(modelsummary_factory_latex = 'kableExtra')
## options(modelsummary_factory_html = 'kableExtra')
##
## Silence this message forever:
##
## config_modelsummary(startup_message = FALSE)##
## Attaching package: 'modelsummary'## The following objects are masked from 'package:DescTools':
##
## Format, Mean, Median, N, SD, Varmodeloalt=list('modeloalt'=modeloalt)
modelsummary(modeloalt, title = "Regresion: modelo alternativo",
stars = TRUE,
output = "kableExtra")| modeloalt | |
|---|---|
| (Intercept) | 0.437*** |
| (0.070) | |
| Puntuación | -0.081 |
| (0.071) | |
| Participaciónpolítica | 0.058** |
| (0.019) | |
| Culturapolítica | 0.027* |
| (0.012) | |
| Derechosciviles | 0.009 |
| (0.039) | |
| Funcionamientodelgobierno | 0.053* |
| (0.019) | |
| Num.Obs. | 20 |
| R2 | 0.730 |
| R2 Adj. | 0.633 |
| AIC | -59.6 |
| BIC | -52.6 |
| Log.Lik. | 36.804 |
| RMSE | 0.04 |
| + p < 0.1, * p < 0.05, ** p < 0.01, *** p < 0.001 |
#Análisis Factorial Exploratorio (EFA)
dontselect=c("País")
select=setdiff(names(Democracia_IDH_AL),dontselect)
theData=Democracia_IDH_AL[,select]
library(magrittr)
head(theData,8)%>%
rmarkdown::paged_table()library(polycor)correlation_matrix <- cor(theData)round(correlation_matrix,2)## Puntuación Procesoelectoralypluralismo
## Puntuación 1.00 0.93
## Procesoelectoralypluralismo 0.93 1.00
## Funcionamientodelgobierno 0.91 0.78
## Participaciónpolítica 0.66 0.67
## Culturapolítica 0.52 0.24
## Derechosciviles 0.98 0.93
## IDH 0.62 0.43
## Funcionamientodelgobierno Participaciónpolítica
## Puntuación 0.91 0.66
## Procesoelectoralypluralismo 0.78 0.67
## Funcionamientodelgobierno 1.00 0.50
## Participaciónpolítica 0.50 1.00
## Culturapolítica 0.51 -0.02
## Derechosciviles 0.85 0.62
## IDH 0.71 0.58
## Culturapolítica Derechosciviles IDH
## Puntuación 0.52 0.98 0.62
## Procesoelectoralypluralismo 0.24 0.93 0.43
## Funcionamientodelgobierno 0.51 0.85 0.71
## Participaciónpolítica -0.02 0.62 0.58
## Culturapolítica 1.00 0.52 0.40
## Derechosciviles 0.52 1.00 0.55
## IDH 0.40 0.55 1.00#Gráfico con todas las variables
ggcorrplot(correlation_matrix)
#viendo que datos permiten factorizar
library(psych)##
## Attaching package: 'psych'## The following object is masked from 'package:polycor':
##
## polyserial## The following object is masked from 'package:modelsummary':
##
## SD## The following objects are masked from 'package:ggplot2':
##
## %+%, alpha## The following objects are masked from 'package:DescTools':
##
## AUC, ICC, SDpsych::KMO(correlation_matrix) ## Kaiser-Meyer-Olkin factor adequacy
## Call: psych::KMO(r = correlation_matrix)
## Overall MSA = 0.43
## MSA for each item =
## Puntuación Procesoelectoralypluralismo
## 0.46 0.43
## Funcionamientodelgobierno Participaciónpolítica
## 0.43 0.31
## Culturapolítica Derechosciviles
## 0.20 0.54
## IDH
## 0.97#viendo si la matriz de correlación es adecuada
cortest.bartlett(correlation_matrix,n=nrow(theData))$p.value>0.05## [1] FALSEResultado: es una matriz singular
fa.parallel(theData, fa = 'fa',correct = T,plot = F)## Warning in fa.stats(r = r, f = f, phi = phi, n.obs = n.obs, np.obs = np.obs, :
## The estimated weights for the factor scores are probably incorrect. Try a
## different factor score estimation method.## Warning in fac(r = r, nfactors = nfactors, n.obs = n.obs, rotate = rotate, : An
## ultra-Heywood case was detected. Examine the results carefully## Warning in fa.stats(r = r, f = f, phi = phi, n.obs = n.obs, np.obs = np.obs, :
## The estimated weights for the factor scores are probably incorrect. Try a
## different factor score estimation method.
## Warning in fa.stats(r = r, f = f, phi = phi, n.obs = n.obs, np.obs = np.obs, :
## The estimated weights for the factor scores are probably incorrect. Try a
## different factor score estimation method.## Warning in fac(r = r, nfactors = nfactors, n.obs = n.obs, rotate = rotate, : An
## ultra-Heywood case was detected. Examine the results carefully## Warning in fa.stats(r = r, f = f, phi = phi, n.obs = n.obs, np.obs = np.obs, :
## The estimated weights for the factor scores are probably incorrect. Try a
## different factor score estimation method.## Warning in fac(r = r, nfactors = nfactors, n.obs = n.obs, rotate = rotate, : An
## ultra-Heywood case was detected. Examine the results carefully## Warning in fa.stats(r = r, f = f, phi = phi, n.obs = n.obs, np.obs = np.obs, :
## The estimated weights for the factor scores are probably incorrect. Try a
## different factor score estimation method.## Warning in fac(r = r, nfactors = nfactors, n.obs = n.obs, rotate = rotate, : An
## ultra-Heywood case was detected. Examine the results carefully## Warning in fa.stats(r = r, f = f, phi = phi, n.obs = n.obs, np.obs = np.obs, :
## The estimated weights for the factor scores are probably incorrect. Try a
## different factor score estimation method.## Warning in fac(r = r, nfactors = nfactors, n.obs = n.obs, rotate = rotate, : An
## ultra-Heywood case was detected. Examine the results carefully## Warning in fa.stats(r = r, f = f, phi = phi, n.obs = n.obs, np.obs = np.obs, :
## The estimated weights for the factor scores are probably incorrect. Try a
## different factor score estimation method.## Warning in fac(r = r, nfactors = nfactors, n.obs = n.obs, rotate = rotate, : An
## ultra-Heywood case was detected. Examine the results carefully## Warning in fa.stats(r = r, f = f, phi = phi, n.obs = n.obs, np.obs = np.obs, :
## The estimated weights for the factor scores are probably incorrect. Try a
## different factor score estimation method.## Warning in fac(r = r, nfactors = nfactors, n.obs = n.obs, rotate = rotate, : An
## ultra-Heywood case was detected. Examine the results carefully## Warning in fa.stats(r = r, f = f, phi = phi, n.obs = n.obs, np.obs = np.obs, :
## The estimated weights for the factor scores are probably incorrect. Try a
## different factor score estimation method.## Warning in fac(r = r, nfactors = nfactors, n.obs = n.obs, rotate = rotate, : An
## ultra-Heywood case was detected. Examine the results carefully## Warning in fa.stats(r = r, f = f, phi = phi, n.obs = n.obs, np.obs = np.obs, :
## The estimated weights for the factor scores are probably incorrect. Try a
## different factor score estimation method.## Warning in fac(r = r, nfactors = nfactors, n.obs = n.obs, rotate = rotate, : An
## ultra-Heywood case was detected. Examine the results carefully## Warning in fa.stats(r = r, f = f, phi = phi, n.obs = n.obs, np.obs = np.obs, :
## The estimated weights for the factor scores are probably incorrect. Try a
## different factor score estimation method.## Warning in fac(r = r, nfactors = nfactors, n.obs = n.obs, rotate = rotate, : An
## ultra-Heywood case was detected. Examine the results carefully## Parallel analysis suggests that the number of factors = 1 and the number of components = NARESULTADO: Dado que las variables son menor a 0.6, no es posible hacer el Análisis Factorial Exploratorio, varimix ni oblimin pueden ser empleadas
#Análisis de conglomerados
Distribución de componentes de la democracia
boxplot(Democracia_IDH_AL[,c(2:7)],horizontal = F,las=2,cex.axis = 0.5)
library(BBmisc)##
## Attaching package: 'BBmisc'## The following object is masked from 'package:DescTools':
##
## %nin%## The following object is masked from 'package:base':
##
## isFALSEboxplot(normalize(Democracia_IDH_AL[,c(2:7)],method='range',range=c(0,10)))
#Correlación
cor(Democracia_IDH_AL[,c(2:8)])## Puntuación Procesoelectoralypluralismo
## Puntuación 1.0000000 0.9302970
## Procesoelectoralypluralismo 0.9302970 1.0000000
## Funcionamientodelgobierno 0.9088101 0.7838096
## Participaciónpolítica 0.6644540 0.6705509
## Culturapolítica 0.5238965 0.2434508
## Derechosciviles 0.9810601 0.9265352
## IDH 0.6249085 0.4285165
## Funcionamientodelgobierno Participaciónpolítica
## Puntuación 0.9088101 0.66445396
## Procesoelectoralypluralismo 0.7838096 0.67055088
## Funcionamientodelgobierno 1.0000000 0.49625015
## Participaciónpolítica 0.4962501 1.00000000
## Culturapolítica 0.5051464 -0.01544892
## Derechosciviles 0.8527415 0.61987591
## IDH 0.7051060 0.57783965
## Culturapolítica Derechosciviles IDH
## Puntuación 0.52389652 0.9810601 0.6249085
## Procesoelectoralypluralismo 0.24345075 0.9265352 0.4285165
## Funcionamientodelgobierno 0.50514637 0.8527415 0.7051060
## Participaciónpolítica -0.01544892 0.6198759 0.5778396
## Culturapolítica 1.00000000 0.5236906 0.4007599
## Derechosciviles 0.52369061 1.0000000 0.5483545
## IDH 0.40075990 0.5483545 1.0000000#Clusterización
dataClus=Democracia_IDH_AL[,c(2:8)]
row.names(dataClus)=Democracia_IDH_AL$País## Warning: Setting row names on a tibble is deprecated.library(cluster)
g.dist = daisy(dataClus, metric="gower")#viendo cantidad de clusters
library(factoextra)## Welcome! Want to learn more? See two factoextra-related books at https://goo.gl/ve3WBafviz_nbclust(dataClus, pam,diss=g.dist,method = "gap_stat",k.max = 8,verbose = F)
RESULTADO: 3 Clusters
#1. PAM
library(kableExtra)
set.seed(123)
res.pam=pam(g.dist,3,cluster.only = F)
Democracia_IDH_AL$pam=res.pam$cluster
head(Democracia_IDH_AL,15)%>%kbl()%>%kable_styling()| País | Puntuación | Procesoelectoralypluralismo | Funcionamientodelgobierno | Participaciónpolítica | Culturapolítica | Derechosciviles | IDH | pam |
|---|---|---|---|---|---|---|---|---|
| Argentina | 7.02 | 9.17 | 5.36 | 6.11 | 6.25 | 8.24 | 0.841 | 1 |
| Bolivia | 5.70 | 7.50 | 4.64 | 5.56 | 3.75 | 7.06 | 0.718 | 1 |
| Brasil | 6.97 | 9.58 | 5.36 | 6.67 | 5.00 | 8.24 | 0.758 | 1 |
| Chile | 7.97 | 9.58 | 8.57 | 4.44 | 8.13 | 9.12 | 0.849 | 2 |
| Colombia | 6.96 | 9.17 | 6.79 | 5.00 | 5.63 | 8.24 | 0.756 | 1 |
| Costa Rica | 8.07 | 9.58 | 7.50 | 6.67 | 7.50 | 9.12 | 0.811 | 2 |
| Cuba | 3.00 | 1.08 | 3.57 | 3.33 | 4.38 | 2.65 | 0.759 | 3 |
| Ecuador | 6.27 | 8.75 | 5.36 | 6.11 | 4.38 | 7.65 | 0.734 | 1 |
| El Salvador | 5.96 | 9.17 | 4.29 | 5.56 | 3.75 | 7.06 | 0.676 | 1 |
| Guatemala | 5.60 | 7.92 | 5.36 | 3.89 | 4.38 | 6.47 | 0.645 | 1 |
| Haití | 4.91 | 5.58 | 2.93 | 3.89 | 6.25 | 5.88 | 0.559 | 3 |
| Honduras | 5.63 | 8.50 | 4.64 | 4.44 | 4.38 | 6.18 | 0.629 | 1 |
| México | 6.19 | 8.33 | 6.07 | 7.22 | 3.13 | 6.18 | 0.781 | 1 |
| Nicaragua | 3.63 | 2.67 | 1.86 | 3.89 | 5.63 | 4.12 | 0.661 | 3 |
| Panamá | 7.05 | 9.58 | 6.07 | 6.67 | 5.00 | 7.94 | 0.809 | 1 |
fviz_silhouette(res.pam,print.summary = F)
silPAM=data.frame(res.pam$silinfo$widths)
silPAM$País=row.names(silPAM)
poorPAM=silPAM[silPAM$sil_width<0,'País']%>%sort()
poorPAM## character(0)#aggregate(.~ pam, data=dataClus,mean)Democracia_IDH_AL$pamIDHpoor=Democracia_IDH_AL$País%in%poorPAM
Democracia_IDH_AL$pamIDH=as.ordered(Democracia_IDH_AL$pam)
Democracia_IDH_AL$pam=NULL#2. AGNES
fviz_nbclust(dataClus, hcut,diss=g.dist,method = "gap_stat",k.max = 8,verbose = F,hc_func = "agnes")
set.seed(123)
library(factoextra)
res.agnes<- hcut(g.dist, k = 3,hc_func='agnes',hc_method = "ward.D")
Democracia_IDH_AL$agnes=res.agnes$cluster
# ver
head(dataClus,20)%>%kbl()%>%kable_styling()| Puntuación | Procesoelectoralypluralismo | Funcionamientodelgobierno | Participaciónpolítica | Culturapolítica | Derechosciviles | IDH |
|---|---|---|---|---|---|---|
| 7.02 | 9.17 | 5.36 | 6.11 | 6.25 | 8.24 | 0.841 |
| 5.70 | 7.50 | 4.64 | 5.56 | 3.75 | 7.06 | 0.718 |
| 6.97 | 9.58 | 5.36 | 6.67 | 5.00 | 8.24 | 0.758 |
| 7.97 | 9.58 | 8.57 | 4.44 | 8.13 | 9.12 | 0.849 |
| 6.96 | 9.17 | 6.79 | 5.00 | 5.63 | 8.24 | 0.756 |
| 8.07 | 9.58 | 7.50 | 6.67 | 7.50 | 9.12 | 0.811 |
| 3.00 | 1.08 | 3.57 | 3.33 | 4.38 | 2.65 | 0.759 |
| 6.27 | 8.75 | 5.36 | 6.11 | 4.38 | 7.65 | 0.734 |
| 5.96 | 9.17 | 4.29 | 5.56 | 3.75 | 7.06 | 0.676 |
| 5.60 | 7.92 | 5.36 | 3.89 | 4.38 | 6.47 | 0.645 |
| 4.91 | 5.58 | 2.93 | 3.89 | 6.25 | 5.88 | 0.559 |
| 5.63 | 8.50 | 4.64 | 4.44 | 4.38 | 6.18 | 0.629 |
| 6.19 | 8.33 | 6.07 | 7.22 | 3.13 | 6.18 | 0.781 |
| 3.63 | 2.67 | 1.86 | 3.89 | 5.63 | 4.12 | 0.661 |
| 7.05 | 9.58 | 6.07 | 6.67 | 5.00 | 7.94 | 0.809 |
| 6.24 | 8.75 | 5.71 | 5.00 | 4.38 | 7.35 | 0.742 |
| 6.60 | 9.17 | 5.00 | 5.56 | 5.63 | 7.65 | 0.758 |
| 6.54 | 9.17 | 5.36 | 6.11 | 5.00 | 7.06 | 0.765 |
| 8.38 | 10.00 | 8.57 | 6.11 | 7.50 | 9.71 | 0.820 |
| 3.16 | 1.67 | 1.79 | 4.44 | 4.38 | 3.53 | 0.691 |
#Aglomeración Agnes
fviz_dend(res.agnes, cex = 0.7, horiz = T,main = "")## Warning: The `<scale>` argument of `guides()` cannot be `FALSE`. Use "none" instead as
## of ggplot2 3.3.4.
## ℹ The deprecated feature was likely used in the factoextra package.
## Please report the issue at <https://github.com/kassambara/factoextra/issues>.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.
fviz_silhouette(res.agnes,print.summary = F)
silAGNES=data.frame(res.agnes$silinfo$widths)
silAGNES$country=row.names(silAGNES)
poorAGNES=silAGNES[silAGNES$sil_width<0,'country']%>%sort()
poorAGNES## [1] "Argentina" "Haití"RESULTADO: Argentina y Haití mal clusterizados
#aggregate(.~ agnes, data=Democracia_IDH_AL,mean)Democracia_IDH_AL$agnesIDHpoor=Democracia_IDH_AL$País%in%poorAGNES
Democracia_IDH_AL$agnesIDH=as.ordered(Democracia_IDH_AL$agnes)
Democracia_IDH_AL$agnes=NULL#3. GAP
fviz_nbclust(dataClus, hcut,diss=g.dist,method = "gap_stat",k.max = 8,verbose = F,hc_func = "diana")
#Clusterización con DIANA
set.seed(123)
res.diana <- hcut(g.dist, k = 3,hc_func='diana')
dataClus$diana=res.diana$cluster
head(dataClus,15)%>%kbl%>%kable_styling()| Puntuación | Procesoelectoralypluralismo | Funcionamientodelgobierno | Participaciónpolítica | Culturapolítica | Derechosciviles | IDH | diana |
|---|---|---|---|---|---|---|---|
| 7.02 | 9.17 | 5.36 | 6.11 | 6.25 | 8.24 | 0.841 | 1 |
| 5.70 | 7.50 | 4.64 | 5.56 | 3.75 | 7.06 | 0.718 | 1 |
| 6.97 | 9.58 | 5.36 | 6.67 | 5.00 | 8.24 | 0.758 | 1 |
| 7.97 | 9.58 | 8.57 | 4.44 | 8.13 | 9.12 | 0.849 | 2 |
| 6.96 | 9.17 | 6.79 | 5.00 | 5.63 | 8.24 | 0.756 | 1 |
| 8.07 | 9.58 | 7.50 | 6.67 | 7.50 | 9.12 | 0.811 | 2 |
| 3.00 | 1.08 | 3.57 | 3.33 | 4.38 | 2.65 | 0.759 | 3 |
| 6.27 | 8.75 | 5.36 | 6.11 | 4.38 | 7.65 | 0.734 | 1 |
| 5.96 | 9.17 | 4.29 | 5.56 | 3.75 | 7.06 | 0.676 | 1 |
| 5.60 | 7.92 | 5.36 | 3.89 | 4.38 | 6.47 | 0.645 | 1 |
| 4.91 | 5.58 | 2.93 | 3.89 | 6.25 | 5.88 | 0.559 | 3 |
| 5.63 | 8.50 | 4.64 | 4.44 | 4.38 | 6.18 | 0.629 | 1 |
| 6.19 | 8.33 | 6.07 | 7.22 | 3.13 | 6.18 | 0.781 | 1 |
| 3.63 | 2.67 | 1.86 | 3.89 | 5.63 | 4.12 | 0.661 | 3 |
| 7.05 | 9.58 | 6.07 | 6.67 | 5.00 | 7.94 | 0.809 | 1 |
fviz_dend(res.diana, cex = 0.7, horiz = T, main = "")
fviz_silhouette(res.diana,print.summary = F)
silDIANA=data.frame(res.diana$silinfo$widths)
silDIANA$Country=row.names(silDIANA)
poorDIANA=silDIANA[silDIANA$sil_width<0,'Country']%>%sort()
poorDIANA## character(0)dataClus$diana=dplyr::recode(dataClus$diana, `1` = 1, `4`=2,`2`=3,`3`=4)#Visualización comparativa
proyeccion = cmdscale(g.dist, k=2,add = T)
head(proyeccion$points,20)## [,1] [,2]
## Argentina 0.21775459 -0.02900172
## Bolivia -0.06829822 0.15457134
## Brasil 0.16616471 0.09121768
## Chile 0.32173007 -0.31295474
## Colombia 0.13540714 -0.05122613
## Costa Rica 0.39006049 -0.12364613
## Cuba -0.48017816 -0.06990417
## Ecuador 0.05864136 0.14306340
## El Salvador -0.04836731 0.15369033
## Guatemala -0.16120426 0.04966532
## Haití -0.30449662 -0.16889724
## Honduras -0.15800121 0.06428661
## México 0.07845949 0.23152622
## Nicaragua -0.46870234 -0.18901788
## Panamá 0.21610114 0.07233221
## Paraguay 0.01287981 0.08841961
## Perú 0.08433386 0.02616666
## R. Dominicana 0.09357619 0.09949757
## Uruguay 0.41997293 -0.17444272
## Venezuela -0.50583364 -0.05534619Democracia_IDH_AL$dim1 <- proyeccion$points[,1]
Democracia_IDH_AL$dim2 <- proyeccion$points[,2]