Univariate Analysis - Numerical

What is Univariate Analysis?

Univariate = one variable. Analyzing one column by itself.

Analyzing Numbers

For numerical columns (age, salary, price):

code.py

import pandas as pd

df = pd.DataFrame({
    'Age': [22, 25, 28, 30, 32, 35, 40, 45, 50, 65]
})

# Basic stats
print(df['Age'].describe())

Key Statistics to Check

code.py

# Center of data
print("Mean:", df['Age'].mean())
print("Median:", df['Age'].median())

# Spread of data
print("Min:", df['Age'].min())
print("Max:", df['Age'].max())
print("Std:", df['Age'].std())

# Quartiles
print("25%:", df['Age'].quantile(0.25))
print("75%:", df['Age'].quantile(0.75))

Check Distribution Shape

code.py

# Skewness: is data tilted left or right?
print("Skewness:", df['Age'].skew())

Skew = 0: Balanced (symmetric)
Skew > 0: Tail on right (most values are low)
Skew < 0: Tail on left (most values are high)

Count Values in Ranges (Bins)

code.py

# How many people in each age group?
bins = [0, 30, 40, 50, 100]
labels = ['Young', 'Middle', 'Senior', 'Old']

df['Age_Group'] = pd.cut(df['Age'], bins=bins, labels=labels)
print(df['Age_Group'].value_counts())

Simple Histogram (Text)

code.py

# Quick view of distribution
print(df['Age'].value_counts(bins=5).sort_index())

Check for Problems

code.py

# Any missing?
print("Missing:", df['Age'].isna().sum())

# Any zeros (if shouldn't be)?
print("Zeros:", (df['Age'] == 0).sum())

# Any negatives?
print("Negative:", (df['Age'] < 0).sum())

# Unique values
print("Unique values:", df['Age'].nunique())

Find Unusual Values

code.py

# Values far from normal
mean = df['Age'].mean()
std = df['Age'].std()

# Values more than 2 std from mean
unusual = df[(df['Age'] < mean - 2*std) | (df['Age'] > mean + 2*std)]
print("Unusual values:", unusual)

Quick Analysis Template

code.py

def analyze_numeric(series):
    print(f"Column: {series.name}")
    print(f"Count: {series.count()}")
    print(f"Missing: {series.isna().sum()}")
    print(f"Mean: {series.mean():.2f}")
    print(f"Median: {series.median():.2f}")
    print(f"Std: {series.std():.2f}")
    print(f"Min: {series.min()}")
    print(f"Max: {series.max()}")
    print(f"Skew: {series.skew():.2f}")

analyze_numeric(df['Age'])

Key Points

Univariate = analyze one column
Check: mean, median, min, max, std
Look at distribution shape (skewness)
Find missing values and outliers
Group into bins to see patterns

What's Next?

Learn to analyze categorical columns (text like city, gender, status).