🏠 Nashville Housing Market Analysis

Advanced SQL Analytics & Market Segmentation (SQL Server)

---

📌 Project Overview

This project is a structured, end-to-end SQL analytics case study analyzing the Nashville housing market using Microsoft SQL Server.

The objective was to transform raw transactional housing data into a clean, analysis-ready dataset and extract meaningful business insights using advanced SQL techniques.

This repository emphasizes SQL-based analytical modeling rather than dashboard visualization.

SQL Depth Demonstrated:

• Structured data cleaning workflows
• Feature engineering
• Window functions (PERCENTILE_CONT, LAG, ROW_NUMBER)
• Market share modeling
• Time-series analysis
• Segmentation logic
• Business interpretation from raw data

---

🏗 Project Architecture

The project follows a structured, production-style workflow:

Raw Dataset
↓
Data Profiling
↓
Data Cleaning
↓
Feature Engineering
↓
Analytical Modeling
↓
Business Insights

Each stage is separated into dedicated SQL files to reflect professional project organization and clear analytical layering.

---

📂 Repository Structure

nashville-housing-sql-analytics/
│
├── dataset/
│ ├── 01_database_setup.sql
│ ├── 02_table_creation.sql
│ ├── 03_data_profiling.sql
│ ├── 04_data_cleaning.sql
│ ├── 05_feature_engineering.sql
│ └── 06_analysis.sql
│
└── README.md

🧹 1️⃣ Data Cleaning

Key transformations performed:

• Standardizing categorical values (SoldAsVacant)
• Converting SalePrice into numeric format
• Imputing missing PropertyAddress values using self-join logic
• Splitting composite address fields into structured columns
• Removing duplicate records using ROW_NUMBER()
• Creating a structured analytical table (nashville_clean)

The resulting dataset is normalized and analysis-ready.

---

🏗 2️⃣ Feature Engineering

Engineered analytical features to support segmentation and time-series modelling:

• SaleYear, SaleMonth, SaleQuarter
• PropertyAgeAtSale
• PropertyAgeGroup classification
• Time-based modeling fields
• Market share calculations

This step transformed raw transactional data into a structured analytical layer suitable for advanced modelling.

---

📊 3️⃣ Time-Based Market Analysis

Performed trend and growth analysis using window functions.

Techniques Used:

• PERCENTILE_CONT for median pricing
• LAG() for Year-over-Year growth modeling
• Partitioned aggregations for yearly summaries

Key Findings:

• Strong expansion phase between 2013–2015
• Pricing growth outpaced volume growth in certain periods
• Post-2015 stabilization reflects cyclical normalization rather than structural collapse
• Median price was used over average due to right-skewed pricing distribution

The market demonstrates cyclical growth behavior consistent with normal economic patterns.

---

🏘 4️⃣ Property Age Segmentation

Segmented properties by lifecycle stage to understand pricing and demand concentration.

Key Findings:

• ~42% of transactions occur in older resale inventory
• New construction and historic homes command pricing premiums
• Pricing dispersion varies significantly across age segments
• Older homes dominate volume, while newer homes dominate premium positioning

The housing market is structurally segmented by property age.

---

📍 5️⃣ Geographic Segmentation

Analyzed transaction volume and pricing behavior across municipalities.

Key Findings:

• Nashville accounts for ~71% of total transaction volume
• Suburban cities (e.g., Nolensville, Brentwood) command higher median prices
• Pricing dispersion varies materially across cities
• Urban core shows wider price spread due to mixed-income and luxury outliers

The market exhibits clear geographic segmentation rather than uniform pricing behavior.

---

🧠 Advanced SQL Techniques Demonstrated

This project showcases strong SQL depth including:

• Common Table Expressions (CTEs)
• Window Functions:
  • ROW_NUMBER()
  • PERCENTILE_CONT()
  • LAG()
• Partitioned aggregations
• Market share modeling using window sums
• Duplicate detection via partition logic
• Time-series growth modeling
• Structured query layering

---

🎯 Business Questions Answered

• Is the housing market growing?
• Is pricing increasing?
• Is transaction volume increasing?
• Which property age segments drive demand?
• Which age segments command pricing premiums?
• Is the market geographically concentrated?
• Does pricing distribution vary across locations?

Each question is answered using structured SQL modeling and analytical reasoning.

---

🏁 Conclusion

This project demonstrates the ability to:

• Transform messy raw data into structured analytical datasets
• Apply statistical reasoning within SQL
• Design production-grade query architecture
• Use window functions effectively
• Derive meaningful business insights from transactional data

This repository serves as a SQL depth showcase, emphasizing analytical thinking and structured query design rather than visualization tools.