LEARN COMPLETE PYTHON IN 24 HOURS

6. Exploratory Data Analysis (EDA) in R

Exploratory Data Analysis (EDA) is the process of investigating a dataset to discover patterns, spot anomalies, test hypotheses, and check assumptions — before building any model. In R, EDA is extremely powerful thanks to tidyverse, ggplot2, and specialized packages.

Core goals of EDA

• Understand data structure & quality

• Identify missing values, outliers, errors

• Discover relationships between variables

• Detect patterns (trend, seasonality, clusters)

• Guide feature engineering and modeling decisions

6.1 Summary Statistics & Descriptive Analysis

Start every EDA with a quick overview of the data.

Basic summary functions

R

library(tidyverse) # Load example dataset data("mtcars") # Quick overview glimpse(mtcars) # structure & types summary(mtcars) # min, max, mean, median, quartiles skimr::skim(mtcars) # very detailed summary (install skimr)

Custom summary by group

R

mtcars %>% group_by(cyl) %>% summarise( avg_mpg = mean(mpg, na.rm = TRUE), median_hp = median(hp), sd_wt = sd(wt), n = n() ) %>% arrange(desc(avg_mpg))

Best practice: Always use na.rm = TRUE and check for missing values first.

6.2 Handling Missing Values & Outliers

Detect missing values

R

# Count missing per column colSums(is.na(airquality)) # Percentage missing colMeans(is.na(airquality)) * 100 # Visual: missingno style (install naniar) library(naniar) vis_miss(airquality) gg_miss_var(airquality)

Handling missing values

R

# 1. Drop rows/columns with missing airquality_complete <- airquality %>% drop_na() airquality %>% drop_na(Ozone) # drop only if Ozone missing # 2. Impute with mean/median airquality %>% mutate(Ozone = if_else(is.na(Ozone), mean(Ozone, na.rm = TRUE), Ozone)) # 3. Impute with last observation carried forward (time series) library(zoo) airquality$Ozone <- na.locf(airquality$Ozone, na.rm = FALSE) # 4. Advanced imputation (missForest, mice packages)

Detecting & handling outliers

R

# Boxplot visual ggplot(airquality, aes(y = Ozone)) + geom_boxplot(fill = "lightblue") + labs(title = "Ozone Outliers") # IQR method Q1 <- quantile(airquality$Ozone, 0.25, na.rm = TRUE) Q3 <- quantile(airquality$Ozone, 0.75, na.rm = TRUE) IQR <- Q3 - Q1 lower <- Q1 - 1.5 IQR upper <- Q3 + 1.5 IQR # Flag outliers airquality <- airquality %>% mutate(ozone_outlier = Ozone < lower | Ozone > upper) # Winsorize (cap) outliers airquality$Ozone_winsor <- pmin(pmax(airquality$Ozone, lower), upper)

Tip: Never blindly remove outliers — investigate first (measurement error? interesting case?).

6.3 Univariate, Bivariate & Multivariate EDA

Univariate (one variable)

R

# Categorical ggplot(diamonds, aes(x = cut)) + geom_bar(fill = "steelblue") + labs(title = "Diamond Cut Distribution") # Numerical ggplot(diamonds, aes(x = price)) + geom_histogram(bins = 50, fill = "coral") + labs(title = "Price Distribution") # Density + boxplot ggplot(diamonds, aes(x = price)) + geom_density(fill = "lightgreen", alpha = 0.5) + geom_boxplot(width = 0.1, fill = "white") + labs(title = "Price Density & Boxplot")

Bivariate (two variables)

R

# Numeric vs Numeric ggplot(mtcars, aes(x = wt, y = mpg)) + geom_point(aes(color = factor(cyl)), size = 3) + geom_smooth(method = "lm", color = "red") + labs(title = "Weight vs MPG by Cylinders") # Categorical vs Numeric ggplot(tips, aes(x = day, y = total_bill, fill = day)) + geom_boxplot() + labs(title = "Total Bill by Day") # Categorical vs Categorical ggplot(tips, aes(x = sex, fill = smoker)) + geom_bar(position = "fill") + labs(title = "Smoking Status by Gender (proportions)")

Multivariate (three or more variables)

R

# Pair plot GGally::ggpairs(tips[, c("total_bill", "tip", "size")], aes(color = smoker)) # Faceted plot ggplot(tips, aes(x = total_bill, y = tip)) + geom_point(aes(color = sex)) + facet_grid(time ~ day) + labs(title = "Tip vs Bill by Time & Day")

6.4 Automated EDA with DataExplorer / SmartEDA

Manual EDA takes time — automated tools generate full reports instantly.

DataExplorer (very popular)

R

# install.packages("DataExplorer") library(DataExplorer) # Generate full EDA report (HTML) create_report(airquality, output_file = "airquality_eda_report.html") # Quick plots plot_intro(airquality) plot_missing(airquality) plot_histogram(airquality) plot_correlation(airquality) plot_boxplot(airquality, by = "Month")

SmartEDA (alternative)

R

# install.packages("SmartEDA") library(SmartEDA) # Target variable analysis (if you have one) ExpReport(airquality, Target = NULL, output_file = "smarteda_report.html")

When to use automated EDA

• First look at new dataset

• Quick report for team/stakeholders

• Identify issues before deep manual analysis

Mini Summary Project – Full EDA on Titanic Dataset

R

library(tidyverse) library(DataExplorer) df <- titanic::titanic_train # 1. Quick overview glimpse(df) create_report(df, output_file = "titanic_eda.html") # 2. Manual key plots ggplot(df, aes(x = Age, fill = Survived)) + geom_histogram(position = "identity", alpha = 0.6, bins = 30) + labs(title = "Age Distribution by Survival") ggplot(df, aes(x = Pclass, fill = factor(Survived))) + geom_bar(position = "fill") + labs(title = "Survival Rate by Passenger Class") # 3. Correlation heatmap (numeric only) df_numeric <- df %>% select(where(is.numeric)) corr <- cor(df_numeric, use = "pairwise.complete.obs") corrplot::corrplot(corr, method = "color", type = "upper", tl.cex = 0.8)

This completes the full Exploratory Data Analysis (EDA) in R section — now you can deeply understand any dataset before modeling or reporting!

📚 Amazon Book Library

All my books are FREE on Amazon Kindle Unlimited🌍 Exclusive Country-Wise Amazon Book Library – Only Here!

On GlobalCodeMaster.com you’ll find complete, ready-to-use lists of my books with direct Amazon links for every country.

Belong to India, Australia, USA, UK, Canada or any other country? Just click your country’s link and enjoy:

✅ Any eBook FREE on Kindle Unlimited ✅ Or buy at incredibly low prices

400+ fresh books written in 2025-2026 with today’s latest AI, Python, Machine Learning & tech trends – nowhere else will you find this complete country-wise collection on one platform!

Choose your country below and start reading instantly 🚀

BOOK LIBRARY USA 2026 LINK

BOOK LIBRARY INDIA 2026 LINK

BOOK LIBRARY AUSTRALIA 2026 LINK

BOOK LIBRARY CANADA 2026 LINK

BOOK LIBRARY UNITED KINGDOM 2026 LINK

BOOK LIBRARY GERMANY 2026 LINK

BOOK LIBRARY FRANCE 2026 LINK

BOOK LIBRARY ITALY 2026 LINK

BOOK LIBRARY SPAIN 2026 LINK

BOOK LIBRARY NETHERLANDS 2026 LINK

BOOK LIBRARY BRAZIL 2026 LINK

BOOK LIBRARY MEXICO 2026 LINK

BOOK LIBRARY JAPAN 2026 LINK

BOOK LIBRARY POLAND 2026 LINK

BOOK LIBRARY IRELAND 2026 LINK

BOOK LIBRARY SWEDEN 2026 LINK

BOOK LIBRARY BELGIUM 2026 LINK