Insurance-Data-Analysis using Python

This project involves exploratory data analysis (EDA) and linear regression modeling on a medical insurance dataset to uncover key insights and predict insurance charges based on individual characteristics.

Dataset

• Source: insurance.csv
• Features:
  • age: Age of the policyholder
  • sex: Gender of the policyholder
  • bmi: Body Mass Index
  • children: Number of dependents
  • smoker: Smoker or non-smoker
  • region: Residential region
  • charges: Insurance cost

Problem Statement

ABC Insurance aims to understand which factors influence medical insurance premiums. The goal is to analyze the dataset and build a model to predict charges using various customer attributes.

Objectives

• Perform EDA to uncover patterns and relationships
• Visualize insights using Python libraries (Matplotlib, Seaborn)
• Build and evaluate a simple linear regression model
• Provide business-level observations

Tools Used

• Python
• Jupyter Notebook
• Pandas, NumPy
• Matplotlib, Seaborn

Key Findings

• Smokers pay significantly higher premiums than non-smokers.
• There is a positive correlation between age, BMI, and charges.
• Southeast region showed relatively higher charges.

Model Summary

A linear regression model was used to predict charges. The model was evaluated using Root Mean Square Error (RMSE) and visualized with predicted vs actual plots.