library(rio)
lima22 = import("Lima2022.xlsx")## New names:
## • `` -> `...2`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, unionlima22 <- mutate(lima22,
ubigeo = as.numeric(gsub("[^0-9]", "", Distrito)),
distri = gsub("[^A-Za-z]", "", Distrito))lima22 = select(lima22, Tasa, ubigeo, distri)#Scrapeo /html/body/div[2]/div/div[3]/main/div[4]/div[3]/div[1]/table[9]
library(rvest)
library(rvest)
link = "https://es.wikipedia.org/wiki/Elecciones_municipales_de_Lima_de_2022"
path = '/html/body/div[2]/div/div[3]/main/div[3]/div[3]/div[1]/table[9]/tbody'
dataWS <- read_html(link)%>%html_nodes(xpath = path)%>%html_table()%>% .[[1]]
head(dataWS)## # A tibble: 6 × 17
## Distrito RP RP PP PP SP SP FE FE APP APP JP
## <chr> <chr> <chr> <chr> <chr> <chr> <chr> <chr> <chr> <chr> <chr> <chr>
## 1 Distrito "" "" "" "" "" "" "" "" "" "" ""
## 2 Distrito "V" "%" "V" "%" "V" "%" "V" "%" "V" "%" "V"
## 3 Ancón "3,725" "13.2… "9,3… "33.… "5,9… "21.… "2,0… "7.2… "3,5… "12.… "1,6…
## 4 Ate "57,374" "17.4… "98,… "29.… "52,… "15.… "27,… "8.5… "26,… "7.9… "25,…
## 5 Barranco "11,604" "36.9… "5,7… "18.… "6,4… "20.… "2,7… "8.8… "2,1… "6.9… "1,6…
## 6 Breña "22,721" "31.1… "18,… "25.… "14,… "19.… "8,2… "11.… "2,4… "3.3… "3,8…
## # ℹ 5 more variables: JP <chr>, AvP <chr>, AvP <chr>, PL <chr>, PL <chr>dataWS <- dataWS[-c(1:2), ]dataWS <- dataWS[-c(44), ]residuos = import("residuosPeru.xlsx")## New names:
## • `` -> `...2`
## • `` -> `...3`
## • `` -> `...4`
## • `` -> `...5`
## • `` -> `...6`
## • `` -> `...7`
## • `` -> `...8`
## • `` -> `...9`
## • `` -> `...10`
## • `` -> `...11`
## • `` -> `...12`
## • `` -> `...13`
## • `` -> `...14`
## • `` -> `...15`colnames(residuos) <- residuos[1,]
residuos <- residuos[-1,]
rownames(residuos) <- NULLresiduos<- subset(residuos, PROVINCIA == "LIMA")colnames(dataWS)[colnames(dataWS) == "Distrito"] <- "DISTRITO"colnames(lima22)[colnames(lima22) == "distri"] <- "DISTRITO"residuos <- residuos %>%
group_by(DISTRITO) %>%
slice_sample(n = 1) %>%
ungroup()remove_accents <- function(x) {
iconv(x, from = "UTF-8", to = "ASCII//TRANSLIT")
}dataWS$DISTRITO <- remove_accents(dataWS$DISTRITO)to_uppercase <- function(x) {
toupper(x)
}dataWS$DISTRITO <- to_uppercase(dataWS$DISTRITO)remove_non_letters <- function(x) {
gsub("[^A-Za-z]", "", x)
}dataWS$DISTRITO <- remove_non_letters(dataWS$DISTRITO)replace_value1 <- function(x) {
sub("LURIGANCHOCHOSICA", "LURIGANCHO", x)
}dataWS$DISTRITO <- replace_value1(dataWS$DISTRITO)residuos$DISTRITO <- remove_non_letters(residuos$DISTRITO)residuos$QRESIDUOS_NO_DOM <- round(as.numeric(residuos$QRESIDUOS_NO_DOM), 2)str(residuos
)## tibble [43 × 15] (S3: tbl_df/tbl/data.frame)
## $ FECHA_CORTE : chr [1:43] "20230614" "20230614" "20230614" "20230614" ...
## $ N_SEC : chr [1:43] "6874" "1264" "5002" "10625" ...
## $ UBIGEO : chr [1:43] "150102" "150103" "150104" "150105" ...
## $ REG_NAT : chr [1:43] "COSTA" "COSTA" "COSTA" "COSTA" ...
## $ DEPARTAMENTO : chr [1:43] "LIMA" "LIMA" "LIMA" "LIMA" ...
## $ PROVINCIA : chr [1:43] "LIMA" "LIMA" "LIMA" "LIMA" ...
## $ DISTRITO : chr [1:43] "ANCON" "ATE" "BARRANCO" "BREA" ...
## $ POB_TOTAL : chr [1:43] "41474" "611082" "29482" "92153" ...
## $ POB_URBANA : chr [1:43] "41474" "611082" "29482" "92153" ...
## $ POB_RURAL : chr [1:43] "0" "0" "0" "0" ...
## $ GPC_DOM : chr [1:43] "0.68" "0.57" "0.78" "0.74" ...
## $ QRESIDUOS_DOM : chr [1:43] "10293.85" "127750" "8415.049999999999" "24890.53" ...
## $ QRESIDUOS_NO_DOM: num [1:43] 4412 54750 3606 10667 31726 ...
## $ QRESIDUOS_MUN : chr [1:43] "14705.5" "182500" "12021.5" "35557.89" ...
## $ PERIODO : chr [1:43] "2017" "2014" "2016" "2019" ...residuos <- residuos %>% select(DISTRITO, QRESIDUOS_NO_DOM)lima22 <- lima22 %>% select(DISTRITO, Tasa)replace_value2 <- function(x) {
sub("MAGDALENAVIEJA", "PUEBLOLIBRE", x)
}residuos$DISTRITO <- replace_value2(residuos$DISTRITO)replace_value3 <- function(x) {
sub("BRE", "BRENA", x)
}lima22$DISTRITO <- replace_value3(lima22$DISTRITO)replace_value4 <- function(x) {
sub("BREA", "BRENA", x)
}residuos$DISTRITO <- replace_value4(residuos$DISTRITO)replace_value5 <- function(x) {
sub("PUEBLOLIBRENA", "PUEBLOLIBRE", x)
}lima22$DISTRITO <- replace_value5(lima22$DISTRITO)dataFinal <- merge(merge(residuos, dataWS, by = "DISTRITO", all = TRUE), lima22, by = "DISTRITO", all = TRUE)dataFinal <- dataFinal %>% select(DISTRITO, Tasa, RP, PP, SP, QRESIDUOS_NO_DOM)cols_to_compare <- dataFinal[, c("RP", "PP", "SP")]max_values <- apply(cols_to_compare, 1, max)cols_to_compare[cols_to_compare == max_values] <- 1
cols_to_compare[cols_to_compare != 1] <- 0dataFinal[, c("RP", "PP", "SP")] <- cols_to_compareh1=formula(RP~QRESIDUOS_NO_DOM + Tasa)str(dataFinal)## 'data.frame': 43 obs. of 6 variables:
## $ DISTRITO : chr "ANCON" "ATE" "BARRANCO" "BRENA" ...
## $ Tasa : chr "113,6" "100,8" "553,7" "218,6" ...
## $ RP : chr "0" "0" "0" "1" ...
## $ PP : chr "1" "1" "0" "0" ...
## $ SP : chr "0" "0" "1" "0" ...
## $ QRESIDUOS_NO_DOM: num 4412 54750 3606 10667 31726 ...dataFinal$Tasa <- gsub(",", ".", dataFinal$Tasa)dataFinal$Tasa <- as.numeric(dataFinal$Tasa)str(dataFinal)## 'data.frame': 43 obs. of 6 variables:
## $ DISTRITO : chr "ANCON" "ATE" "BARRANCO" "BRENA" ...
## $ Tasa : num 114 101 554 219 111 ...
## $ RP : chr "0" "0" "0" "1" ...
## $ PP : chr "1" "1" "0" "0" ...
## $ SP : chr "0" "0" "1" "0" ...
## $ QRESIDUOS_NO_DOM: num 4412 54750 3606 10667 31726 ...dataFinal$RP <- as.factor(dataFinal$RP)rlog = glm(h1, data=dataFinal, family = binomial)summary(rlog)##
## Call:
## glm(formula = h1, family = binomial, data = dataFinal)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -7.359e-01 6.668e-01 -1.104 0.270
## QRESIDUOS_NO_DOM -1.023e-05 1.557e-05 -0.657 0.511
## Tasa 1.724e-03 2.780e-03 0.620 0.535
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 55.618 on 42 degrees of freedom
## Residual deviance: 54.828 on 40 degrees of freedom
## AIC: 60.828
##
## Number of Fisher Scoring iterations: 4library(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')
##
## Change the default backend persistently:
##
## config_modelsummary(factory_default = 'gt')
##
## Silence this message forever:
##
## config_modelsummary(startup_message = FALSE)modelrl=list('Gana RP (I)'=rlog)
modelsummary(modelrl,
title = "Regresión Logística",
exponentiate = T,
stars = TRUE,
output = "kableExtra")| Gana RP (I) | |
|---|---|
| (Intercept) | 0.479 |
| (0.319) | |
| QRESIDUOS_NO_DOM | 1.000 |
| (0.000) | |
| Tasa | 1.002 |
| (0.003) | |
| Num.Obs. | 43 |
| AIC | 60.8 |
| BIC | 66.1 |
| Log.Lik. | -27.414 |
| F | 0.373 |
| RMSE | 0.47 |
| + p < 0.1, * p < 0.05, ** p < 0.01, *** p < 0.001 |
library(margins)
library(kableExtra)##
## Attaching package: 'kableExtra'## The following object is masked from 'package:dplyr':
##
## group_rowsmarginalsData=summary(margins(rlog))
marginalsData%>% kable(caption = "Efectos Marginales Promedio (AME)- Modelo I") %>%kableExtra::kable_styling(full_width = T)| factor | AME | SE | z | p | lower | upper |
|---|---|---|---|---|---|---|
| QRESIDUOS_NO_DOM | -0.0000023 | 0.0000034 | -0.6674578 | 0.5044798 | -0.0000090 | 4.40e-06 |
| Tasa | 0.0003846 | 0.0006099 | 0.6306042 | 0.5282994 | -0.0008108 | 1.58e-03 |
Como los intervalos de confianza pasan por el 0, se evidencia que no tiene un efecto estadísticamente significativo. Asimismo, el modelo en general tiene un pvalue de 0.3, el cual es mucho mayor a 0.05 por lo que es modelo no tiene un efecto estadísticamente significativo. Pero, no se puede descartar totalmente que tenga efecto (?)