Statistics & Mathematics for Data Analysis

A project demonstrating the mathematical foundations of data analysis through exploratory statistics, linear algebra, dimensionality reduction, linear regression, and Bayesian inference.

Project Overview

Completed as part of the Hamrah Aval Data Analysis Bootcamp, this project uses a real advertising dataset to connect statistical concepts with practical data analysis.

Objectives

• Exploratory Data Analysis (EDA)
• Descriptive statistics
• Data visualization
• Covariance matrix analysis
• Singular Value Decomposition (SVD)
• Linear regression from first principles
• Bayesian inference

Workflow

1. Explore the advertising dataset
2. Compute descriptive statistics
3. Visualize feature distributions
4. Analyze covariance relationships
5. Apply Singular Value Decomposition (SVD)
6. Build a linear regression model using linear algebra
7. Evaluate regression coefficients
8. Estimate the probability of high sales using Bayesian inference

Technologies

• Python
• NumPy
• Pandas
• Matplotlib
• SciPy
• Jupyter Notebook

Repository Structure

statistics-mathematics-for-data-analysis/
├── assets/
├── statistics_math_analysis.ipynb
├── project_description.pdf
├── README.md
└── requirements.txt

Skills Demonstrated

• Descriptive Statistics
• Linear Algebra
• Singular Value Decomposition
• Linear Regression
• Bayesian Statistics
• Data Visualization
• Exploratory Data Analysis

Author

Torkan Parvin

GitHub: https://github.com/torkan-parvin LinkedIn: https://www.linkedin.com/in/torkan-parvin/