Statistical Testing & A/B Analysis Project

Project Overview

This project demonstrates the application of multiple statistical tests to analyze business scenarios and make data-driven decisions.

It includes:

• A/B Testing (Z-Test)
• T-Test
• Chi-Square Test
• ANOVA Test

---

Objective

To understand how different statistical tests are applied in real-world business problems such as marketing campaigns, customer behavior, and group comparisons.

---

Statistical Tests Covered

1️⃣ Z-Test (A/B Testing)

• Compared old vs new marketing campaigns
• Evaluated conversion rate differences

2️⃣ T-Test

• Compared means between two groups
• Example: Average spending of two customer segments

3️⃣ Chi-Square Test

• Tested relationship between categorical variables
• Example: Customer response vs campaign type

4️⃣ ANOVA Test

• Compared means across multiple groups
• Example: Sales across different regions or categories

---

Tech Stack

• Python
• Pandas
• NumPy
• Statsmodels / Scipy
• Matplotlib / Seaborn

---

Methodology

1. Data Cleaning & Preprocessing
2. Exploratory Data Analysis (EDA)
3. Applying Statistical Tests
4. Interpreting P-values
5. Drawing Business Conclusions

---

Key Insights

• Different tests solve different types of problems
• Choosing the right statistical test is crucial
• P-value helps validate business decisions

---

Future Improvements

• Add Machine Learning model for prediction
• Build dashboard for visualization
• Automate statistical testing pipeline

---

Author

Nikhil Varkute