📁 Data Manipulation

🔄 Load and preview data

# Import pandas and read in a CSV file
import pandas as pd
df = pd.read_csv("data.csv")

# View the first 5 rows
df.head()

✏️ Rename columns

# Rename a column from 'oldName' to 'new_name'
df.rename(columns={'oldName': 'new_name'}, inplace=True)

🔍 Filter data

# Filter rows where the value column is greater than 10
filtered = df[df['value'] > 10]

📊 Group and aggregate

# Group by category and compute the mean of 'value'
grouped = df.groupby('category')['value'].mean()

🔗 Merge datasets

# Merge df1 and df2 on the 'id' column using inner join
merged = pd.merge(df1, df2, on='id', how='inner')

🕓 Convert to datetime

# Convert a string column to datetime format
df['date'] = pd.to_datetime(df['date'])

🧼 Fill missing values

# Replace missing values in a column with 'Unknown'
df['column'].fillna('Unknown', inplace=True)

📁 Statistical Reference

📏 T-test: compare two independent groups

from scipy.stats import ttest_ind

# Compare means between group_a and group_b
ttest_ind(group_a, group_b)

🧮 Chi-square test: categorical relationships

from scipy.stats import chi2_contingency
import pandas as pd

# Test relationship between two categorical variables
chi2_contingency(pd.crosstab(df['group'], df['outcome']))

🔗 Correlation matrix

# Show correlations between all numeric columns
df.corr()

📉 Linear regression with statsmodels

import statsmodels.api as sm

# Define X (independent vars) and y (target var)
X = df[['var1', 'var2']]
X = sm.add_constant(X)
y = df['target']

# Build and fit the model
model = sm.OLS(y, X).fit()

# Print full summary
print(model.summary())

📊 Confidence interval for a mean

import numpy as np
import scipy.stats as st

# Compute 95% confidence interval for a sample mean
ci = st.t.interval(0.95, len(data)-1, loc=np.mean(data), scale=st.sem(data))

📁 Marketing Calculations

💡 Conversion Rate

# Conversion rate = conversions / visitors
conversions = 125
visitors = 5000
conversion_rate = conversions / visitors

💰 Cost Per Acquisition (CPA)

# CPA = total spend / total acquisitions
spend = 15000
acquisitions = 300
cpa = spend / acquisitions

📈 Customer Lifetime Value (CLV)

# CLV = avg purchase × purchase frequency × avg lifespan
avg_purchase = 75
purchase_freq = 6
avg_customer_lifespan = 3
clv = avg_purchase * purchase_freq * avg_customer_lifespan

📬 Email Open Rate

# Open rate = opened emails / delivered emails
opened = 400
delivered = 1000
open_rate = opened / delivered

📊 Marketing Lift

# Lift = (treatment conversion - control conversion) / control conversion
control = 0.08
treatment = 0.11
lift = (treatment - control) / control

💬 Want to Contribute?

Have a go-to snippet or a favorite stat trick? Drop me a line.