This project analyzes the growth of solar energy generation in the United States using a PESTEL framework (Political, Economic, Environmental) and applies time-series forecasting (SARIMA) to predict future trends.
The objective is to identify key drivers of solar energy expansion and provide data-driven strategic insights for decision-makers.
-
Analyze historical solar energy generation (2015–2024)
-
Identify key external drivers (PESTEL factors)
-
Explore relationships between economic and environmental variables
-
Forecast solar generation for 2025
-
Provide strategic recommendations
Frequency: Monthly
Time Range: 2015–2024
Unit of Analysis: United States
Key Variables:
-
Solar Electricity Generation (MW)
-
Investment (Billion USD)
-
Electricity Price (USD/kWh)
-
CO₂ Emissions
-
Temperature (°F)
-
Precipitation (mm)
- Data Preparation
-
Data cleaning and transformation
-
Time index creation
-
Handling missing values and outliers
- Exploratory Data Analysis (EDA)
-
Distribution analysis
-
Trend visualization
-
Seasonality detection
- Statistical Analysis
-
Correlation analysis
-
Regression modeling
- Time-Series Modeling
-
Stationarity testing (ADF test)
-
ARIMA model
-
SARIMA model (main model)
The SARIMA model was selected because solar generation exhibits:
-
Strong upward trend
-
Clear seasonal cycles
2025 (12 months)
-
Solar energy generation has grown significantly over the past decade
-
Strong seasonal patterns (summer peaks, winter declines)
-
Investment and electricity prices are key drivers of growth
-
Environmental conditions influence solar output
-
Solar expansion contributes to emissions reduction
-
National-level aggregation (no regional breakdown)
-
Limited time horizon
-
External shocks not explicitly modeled
-
Invest in grid infrastructure to support future capacity
-
Maintain policy incentives for renewable energy
-
Expand energy storage solutions
-
Improve climate-based forecasting systems
An interactive dashboard was developed in Power BI to visualize:
-
Solar generation trends
-
PESTEL drivers
-
Forecast scenarios
📁 File: /dashboard/Solar_PESTEL_Dashboard.pbix
Python (Pandas, NumPy, Matplotlib, Seaborn)
Statsmodels (ARIMA, SARIMA)
Power BI
Jupyter Notebook
📂 How to Run
Clone the repository:
git clone https://github.com/ChrisMoises/UNF-Group-Assignment-.git
Install dependencies:
pip install -r requirements.txt
Run the notebook:
jupyter notebook Solar-PESTEL-Forecasting.ipynb
Cristhian Moises Martínez Alay
Daniel Olmedo Zapata Gaibor
María Alejandra Boada Rodríguez
Viviana Rivera Lozano
Yovanni Rojas Cardona