# 🌍 Geographic Analysis of Insurance Claims and Risk Concentration in Kenya
## 📜 Overview
This project analyzes the **geographic distribution of insurance claims across Kenya** to uncover patterns in claim frequency, claim amounts, and regional risk hotspots.
By combining **SQL-based analysis** with **interactive Power BI dashboards**, the study helps insurance companies make **data-driven decisions** in underwriting, pricing, and claims management.
---
## 🎯 Objectives
### Primary Objective
To analyze and visualize the geographic distribution of insurance claims in Kenya.
### Specific Objectives
1. Identify towns/regions with the highest number and amount of claims.
2. Analyze seasonal claim trends (e.g., rainy vs. dry season).
3. Compare claim reasons across different regions.
4. Evaluate average claim severity by region and policy type.
5. Develop interactive dashboards for geographic and temporal analysis.
---
## 📊 Dataset Description
The dataset contains simulated records of **policyholders**, **policies**, and **claims** from various Kenyan towns and regions.
| Table | Description | Rows | Key Columns |
|--------|--------------|-------|--------------|
| `policyholders` | Customer demographics | 1000+ | `policyholder_id`, `name`, `region`, `date_of_birth` |
| `policies` | Policy details | 1200+ | `policy_id`, `policyholder_id`, `start_date`, `end_date`, `premium_amount` |
| `claims` | Claim details | 2000+ | `claim_id`, `policy_id`, `claim_date`, `claim_amount`, `claim_status`, `claim_type` |
📅 **Date Range:** 2018 – 2025
---
## 🛠 Tools & Technologies
- 🐬 **MySQL 8** — Data storage and SQL analysis
- 📊 **Power BI** — Dashboard creation and visualization
- 📄 **Google Sheets** — Data exploration and documentation
---
## 📈 Methodology
1. **Data Preparation**
- Loaded datasets into MySQL
- Cleaned NULL values, corrected invalid dates, standardized region names
2. **Exploratory Data Analysis (SQL)**
- Aggregated claims by region, month, and policy ty …