Histogramas punto 2
ggplot(Telco_Cusomer_Churn, aes(x = MonthlyCharges)) +
geom_histogram(fill = "tan") +
labs(title = "Histograma: Cargos mensuales", x = "MonthlyCharges", y = "Frecuencia")
## `stat_bin()` using `bins = 30`. Pick better value `binwidth`.
ggplot(Telco_Cusomer_Churn, aes(x = TotalCharges)) +
geom_histogram(fill = "blue") +
labs(title = "Histograma: Cargos totales (TotalCharges)", x = "TotalCharges", y = "Frecuencia")
## `stat_bin()` using `bins = 30`. Pick better value `binwidth`.
## Warning: Removed 11 rows containing non-finite outside the scale range
## (`stat_bin()`).
ggplot(Telco_Cusomer_Churn, aes(x = tenure)) +
geom_histogram(fill = "green") +
labs(title = "Histograma: Antigüedad (meses)", x = "tenure (meses)", y = "Frecuencia")
## `stat_bin()` using `bins = 30`. Pick better value `binwidth`.
# Distribución de clientes por Rotación
ggplot(Telco_Cusomer_Churn, aes(x = Churn)) +
geom_bar(fill = "gold") +
labs(title = "Conteo por Rotación (Churn)", x = "Churn", y = "Conteo")
panel grafico
## `stat_bin()` using `bins = 30`. Pick better value `binwidth`.
## `stat_bin()` using `bins = 30`. Pick better value `binwidth`.
## Warning: Removed 11 rows containing non-finite outside the scale range
## (`stat_bin()`).
r = cor(Telco_Cusomer_Churn[, c("tenure","MonthlyCharges","TotalCharges","SeniorCitizen")],
use = "pairwise.complete.obs")
corrplot(r, method = "number")
Diagrama de cajas por (comparativos)
# Punto 4: MonthlyCharges ~ Churn
ggplot(Telco_Cusomer_Churn, aes(y = MonthlyCharges, x = Churn, fill = Churn))+
geom_boxplot() +
labs(title = "Cargos mensuales por Rotación", x = "Churn", y = "MonthlyCharges")
# Punto 5: TotalCharges ~ Churn
ggplot(Telco_Cusomer_Churn, aes(x = Churn, y = TotalCharges, fill = Churn)) +
geom_boxplot() +
labs(title = "Cargos totales por Rotación", x = "Churn", y = "TotalCharges")
## Warning: Removed 11 rows containing non-finite outside the scale range
## (`stat_boxplot()`).
# Extra
ggplot(Telco_Cusomer_Churn, aes(x = Churn, y = tenure, fill = Churn)) +
geom_boxplot() +
labs(title = "Antigüedad (tenure) por Rotación", x = "Churn", y = "tenure (meses)")
# Extra
ggplot(Telco_Cusomer_Churn, aes(x = Contract, y = MonthlyCharges, fill = Contract)) +
geom_boxplot() +
labs(title = "Cargos mensuales por tipo de contrato", x = "Contract", y = "MonthlyCharges")
# Punto 6 (apiladas): Churn (X) con fill = gender
ggplot(Telco_Cusomer_Churn, aes(x = Churn, fill = gender)) +
geom_bar() +
labs(title = "Barras apiladas: Churn por Género", x = "Churn", y = "Conteo", fill = "Género")
# Punto 7: Dispersión + recta (tenure vs MonthlyCharges)
ggplot(Telco_Cusomer_Churn, aes(x = tenure, y = MonthlyCharges)) +
geom_jitter() +
geom_smooth(method = "lm", colour = "red") +
labs(title = "tenure vs MonthlyCharges", x = "tenure (meses)", y = "MonthlyCharges")
## `geom_smooth()` using formula = 'y ~ x'
##Correlación Pearson tenure vs MonthlyCharges
cor(Telco_Cusomer_Churn$tenure,
Telco_Cusomer_Churn$MonthlyCharges,
use = "pairwise.complete.obs")
## [1] 0.2478999
cor.test(Telco_Cusomer_Churn$tenure,
Telco_Cusomer_Churn$MonthlyCharges)
##
## Pearson's product-moment correlation
##
## data: Telco_Cusomer_Churn$tenure and Telco_Cusomer_Churn$MonthlyCharges
## t = 21.472, df = 7041, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.2258523 0.2696936
## sample estimates:
## cor
## 0.2478999
## Modelo lineal simple: MonthlyCharges ~ tenure
modelo <- lm(MonthlyCharges ~ tenure, data = Telco_Cusomer_Churn)
summary(modelo) # R^2, p-valores, etc.
##
## Call:
## lm(formula = MonthlyCharges ~ tenure, data = Telco_Cusomer_Churn)
##
## Residuals:
## Min 1Q Median 3Q Max
## -57.498 -27.251 6.245 24.943 54.376
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 54.92978 0.57476 95.57 <2e-16 ***
## tenure 0.30372 0.01415 21.47 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 29.15 on 7041 degrees of freedom
## Multiple R-squared: 0.06145, Adjusted R-squared: 0.06132
## F-statistic: 461 on 1 and 7041 DF, p-value: < 2.2e-16
coef(modelo) # β0 (intercepto) y β1 (pendiente)
## (Intercept) tenure
## 54.9297785 0.3037246
# Extra
ggplot(Telco_Cusomer_Churn, aes(x = MonthlyCharges, y = TotalCharges)) +
geom_jitter() +
geom_smooth(method = "lm", colour = "blue") +
labs(title = "MonthlyCharges vs TotalCharges", x = "MonthlyCharges", y = "TotalCharges")
## `geom_smooth()` using formula = 'y ~ x'
## Warning: Removed 11 rows containing non-finite outside the scale range
## (`stat_smooth()`).
## Warning: Removed 11 rows containing missing values or values outside the scale range
## (`geom_point()`).
# Extra
ggplot(Telco_Cusomer_Churn, aes(x = tenure, y = TotalCharges)) +
geom_jitter() +
geom_smooth(method = "lm", colour = "tan") +
labs(title = "tenure vs TotalCharges", x = "tenure (meses)", y = "TotalCharges")
## `geom_smooth()` using formula = 'y ~ x'
## Warning: Removed 11 rows containing non-finite outside the scale range
## (`stat_smooth()`).
## Warning: Removed 11 rows containing missing values or values outside the scale range
## (`geom_point()`).
### punto 8
p4 = ggplot(Telco_Cusomer_Churn, aes(y = MonthlyCharges, x = Churn, fill = Churn))+
geom_boxplot()+labs(title="P4: MonthlyCharges ~ Churn", x="Churn", y="MonthlyCharges")
p5 = ggplot(Telco_Cusomer_Churn, aes(x = Churn, y = TotalCharges, fill = Churn))+
geom_boxplot()+labs(title="P5: TotalCharges ~ Churn", x="Churn", y="TotalCharges")
p6 = ggplot(Telco_Cusomer_Churn, aes(x = Churn, fill = gender))+
geom_bar()+labs(title="P6: Churn × Género", x="Churn", y="Conteo", fill="Género")
p7 = ggplot(Telco_Cusomer_Churn, aes(x = tenure, y = MonthlyCharges))+
geom_jitter()+geom_smooth(method="lm", colour="red")+
labs(title="P7: tenure vs MonthlyCharges", x="tenure (meses)", y="MonthlyCharges")
gridExtra::grid.arrange(p4, p5, p6, p7, ncol = 2)
## Warning: Removed 11 rows containing non-finite outside the scale range
## (`stat_boxplot()`).
## `geom_smooth()` using formula = 'y ~ x'
