Sign in with Google
Exploratory Data Analysis (EDA)
Join our community on Telegram!
Join the biggest community of Pharma students and professionals.
Exploratory Data Analysis, commonly known as EDA, is the process of examining and understanding a dataset before applying formal statistical methods or building models. The main goal of EDA is to discover patterns, detect anomalies, check assumptions, and gain insights into the structure of the data.
EDA involves using summary statistics, visualizations, and simple data manipulation techniques to explore the dataset. This step helps analysts identify data quality issues such as missing values, outliers, or inconsistent entries.
In R, EDA is often performed using a combination of base functions, the dplyr package for data manipulation, and the ggplot2 package for visualization.
library(dplyr)
library(ggplot2)One of the first steps in EDA is to inspect the structure of the dataset.
# View structure of dataset
str(mtcars)
# View summary statistics
summary(mtcars)The str() function displays the structure of the dataset, including variable types and sample values. The summary() function provides basic descriptive statistics for each variable.
EDA also involves checking for missing values.
# Count missing values
colSums(is.na(mtcars))Visualizations are an important part of EDA because they help reveal patterns and relationships in the data.
# Histogram
ggplot(mtcars, aes(x = mpg)) +
geom_histogram()
# Scatter plot
ggplot(mtcars, aes(x = wt, y = mpg)) +
geom_point()Another useful EDA step is grouping and summarizing data.
mtcars %>%
group_by(cyl) %>%
summarise(avg_mpg = mean(mpg))Exploratory Data Analysis is a crucial step in any dat
