📱 iPhone Resale Market Intelligence (2026)

An exploratory data analysis and machine learning project investigating the US iPhone resale market post-iPhone 17 launch. This project leverages real marketplace data to identify pricing dynamics, dead inventory, and the key drivers of consumer demand.

📊 Dataset Overview

Dataset sourced from Kaggle.

This project analyzes a unique dataset capturing active iPhone resale listings from publicly available e-commerce platforms in the USA in 2026.

Scope & Details:

• Timeframe: 2026 (Post-iPhone 17 launch window).
• Generations: iPhone 12 through iPhone 17.
• Variants: 23 distinct model configurations (Pro, Pro Max, Plus, Mini).
• Storage: Full range from 64GB to 1TB.
• Nature: Real marketplace listings with actual transaction pricing—not synthetic or estimated data.

Unique Value: Unlike standard pricing datasets, this includes a rare demand signal (units_sold), revealing not just what sellers are asking, but what buyers are actually paying for. This makes it uniquely suited for resale market analysis, depreciation modeling, and consumer electronics pricing research.

🛠️ Project Structure

.

├── demand_analysis.py # Main Python script for analysis and modeling

├── ecommerce_iphone_resale_market_intelligence_usa_2026.csv # Source data

├── price_elasticity.png # Generated: Price vs. Demand analysis

├── demand_drivers.png # Generated: Feature importance chart

└── README.md # This file

🔍 Key Features & Analysis

The demand_analysis.py script performs the following operations:

1. Data Cleaning & Engineering:

   • Cleans pricing data (handling international number formats).
   • Calculates Price Z-Scores to identify over/underpriced listings relative to identical models.
   • Calculates Supply Saturation (count of identical listings currently active).
   • Derives device Age based on generation number.

2. Dead Inventory Detection:

   • Identifies "stale" listings (0 sales) in markets with high supply.
   • Finding: Identified iPhone 14 (128GB) and iPhone 12 (64GB) as the highest risk inventory.

3. Price Elasticity:

   • Analyzes how price tiers impact unit sales for specific high-volume models (e.g., iPhone 13 128GB).

4. Predictive Modeling:

   • Target: is_high_demand (Top 25th percentile of sales).
   • Models: Logistic Regression vs. Random Forest Classifier.
   • Outcome: Random Forest achieved ~71% accuracy in identifying high-demand listings based on price, age, and supply.

🚀 Installation & Usage

Prerequisites

• Python 3.8+
• Virtual Environment (recommended)

Setup

1. Clone the repository

   git clone https://github.com/siyamthanda-code/iphone_classification.git

   cd iphone_classification

2. Create and activate virtual environment (Windows)

   python -m venv venv

   venv\Scripts\activate

3. Install dependencies

   pip install pandas numpy scikit-learn matplotlib seaborn

Running the Analysis

Ensure the dataset file is in the root directory, then run:

python demand_analysis.py

The script will output classification reports to the console and generate two visualization files (price_elasticity.png and demand_drivers.png).

📈 Key Findings

• Supply Glut: The standard iPhone 14 (128GB) is currently the most saturated segment with the highest volume of unsold inventory.
• Demand Drivers: Age and supply saturation are the strongest negative predictors of demand, while competitive pricing (low Z-score) is the strongest positive predictor.
• Model Performance: The Random Forest model significantly outperformed Logistic Regression, suggesting non-linear relationships between price and demand in the resale market.