1. US Car Price Data Analysis

Install Required Libraries

required_packages <- c("ggplot2", "dplyr", "factoextra")
new_packages <- required_packages[!(required_packages %in% installed.packages()[,"Package"])]
if(length(new_packages)) install.packages(new_packages)
lapply(required_packages, library, character.only = TRUE)
## 
## Adjuntando el paquete: 'dplyr'
## The following objects are masked from 'package:stats':
## 
##     filter, lag
## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union
## Welcome! Want to learn more? See two factoextra-related books at https://goo.gl/ve3WBa
## [[1]]
## [1] "ggplot2"   "stats"     "graphics"  "grDevices" "utils"     "datasets" 
## [7] "methods"   "base"     
## 
## [[2]]
## [1] "dplyr"     "ggplot2"   "stats"     "graphics"  "grDevices" "utils"    
## [7] "datasets"  "methods"   "base"     
## 
## [[3]]
##  [1] "factoextra" "dplyr"      "ggplot2"    "stats"      "graphics"  
##  [6] "grDevices"  "utils"      "datasets"   "methods"    "base"

Data Import and Preparation

car_data <- read.csv("C:/Users/Manuel/Desktop/carprice.csv", sep = ",", header = TRUE)

The variable “Type” (column 2) is removed as it is not necessary for further analysis.

car_data_clean <- car_data[, -c(2)]

We verify that only numerical values remain in the relevant columns.

numeric_columns <- car_data_clean[, c("Min.Price", "Price", "Max.Price", "Range.Price", "gpm100", "MPG.city", "MPG.highway")]

Data Normalization and Clustering

Data normalization is performed to scale variables, followed by the computation of the Euclidean distance.

scaled_data <- scale(numeric_columns, center = FALSE, scale = TRUE)
distance_matrix <- dist(scaled_data)

Hierarchical clustering is applied using the complete linkage method.

cluster_complete <- hclust(distance_matrix, method = "complete")
plot(cluster_complete, main = "Car Clustering")

Determining Optimal Number of Clusters

The optimal number of clusters is determined using the elbow method.

ncluster <- fviz_nbclust(numeric_columns, kmeans, method = "wss")
ncluster

The optimal number of clusters is identified as 3 or 4.

num_clusters <- 3
hc_clusters <- hclust(distance_matrix)
plot(as.dendrogram(hc_clusters))
rect.hclust(hc_clusters, k = num_clusters)

Cluster Summary and Insights

A summary of clusters is generated to analyze their characteristics.

cluster_membership <- cutree(hc_clusters, k = num_clusters)
car_data$Cluster <- cluster_membership
cluster_summary <- aggregate(. ~ Cluster, data = car_data[, c("Cluster", "Min.Price", "Price", "Max.Price", "Range.Price", "gpm100", "MPG.city", "MPG.highway")], FUN = mean)
print(cluster_summary)
##   Cluster Min.Price    Price Max.Price Range.Price   gpm100 MPG.city
## 1       1  11.80714 13.76786  15.75357    3.946429 3.846429 23.10714
## 2       2  25.43333 26.44667  27.46667    2.033333 4.560000 17.93333
## 3       3  16.32000 21.86000  27.40000   11.080000 4.780000 18.00000
##   MPG.highway
## 1    29.92857
## 2    26.00000
## 3    24.60000

Conclusion: The clustering analysis of car price data reveals three distinct groups, each representing different segments of the market. Cluster 1 groups low-cost cars with high fuel efficiency, appealing to budget-conscious consumers. Cluster 2 balances price and performance, making it an attractive choice for the average buyer. Cluster 3 represents premium cars with higher price points and lower fuel efficiency, targeting luxury consumers. Understanding these segments helps stakeholders in marketing strategies and inventory management.

2. Canadian Abortion Data Analysis

Install Required Libraries

required_packages <- c("ggplot2", "dplyr", "factoextra")
new_packages <- required_packages[!(required_packages %in% installed.packages()[,"Package"])]
if(length(new_packages)) install.packages(new_packages)
lapply(required_packages, library, character.only = TRUE)
## [[1]]
##  [1] "factoextra" "dplyr"      "ggplot2"    "stats"      "graphics"  
##  [6] "grDevices"  "utils"      "datasets"   "methods"    "base"      
## 
## [[2]]
##  [1] "factoextra" "dplyr"      "ggplot2"    "stats"      "graphics"  
##  [6] "grDevices"  "utils"      "datasets"   "methods"    "base"      
## 
## [[3]]
##  [1] "factoextra" "dplyr"      "ggplot2"    "stats"      "graphics"  
##  [6] "grDevices"  "utils"      "datasets"   "methods"    "base"

Data Preparation

canada_data <- read.csv("C:/Users/Manuel/Desktop/CES11.csv", sep = ",", header = TRUE)
canada_clean <- canada_data[, c("id", "province", "population", "gender", "abortion", "importance", "education", "urban")]

Dummy variables are created for categorical variables.

canada_clean$education_bachelors <- ifelse(canada_clean$education == "bachelors", 1, 0)
canada_clean <- na.omit(canada_clean)

Cluster Interpretation

  • Cluster 1: Predominantly urban population with higher education.
  • Cluster 2: Balanced gender ratio with moderate education.
  • Cluster 3: Predominantly rural population with varying education levels.

Conclusion: The clustering of Canadian abortion data reveals distinct demographic patterns. Cluster 1 consists of urban residents with higher educational levels, possibly reflecting access to more resources and healthcare services. Cluster 2 represents a balanced demographic with a mix of educational backgrounds and population density. Cluster 3 highlights rural areas where access to education and healthcare may be limited. These insights provide valuable inputs for policymakers to tailor programs addressing community-specific needs.