This project analyzes historical data on electric vehicle registrations and provides a forecast for future registrations. Using time series analysis techniques, including SARIMA and Holt-Winters, this project aims to estimate EV adoption trends based on existing data.
- Data Loading and Preprocessing: Load, clean, and prepare EV registration data for analysis.
- Exploratory Data Analysis (EDA): Visualize key trends, distributions, and geographical insights.
- Time Series Analysis and Forecasting: Analyze trends over time and forecast future registrations using SARIMA and other models.
- Data Cleaning: Handles missing values, duplicates, and filters data by completed years up to 2023.
- Exploratory Data Analysis (EDA):
- EV adoption over time
- Distribution by electric vehicle type, manufacturer, and model
- Geographical distribution at county and city levels
- Time Series Forecasting:
- SARIMA: Seasonal Autoregressive Integrated Moving Average model to capture seasonality and trends.
- Alternative Model: Holt-Winters Exponential Smoothing as an option if SARIMA or Prophet is unavailable.
Ensure you have the following libraries installed:
pip install pandas matplotlib seaborn statsmodels
This README covers all main aspects of the project, guiding users through setup, usage, and forecasting steps. Let me know if you’d like additional customization!