Energy Data Business Project

Project Overview

This project focuses on analyzing and forecasting energy production using time series models. The analysis aims to explore trends in energy production by source and producer, identify shifts towards renewable energy, and provide predictions for future energy generation.

Project Members

• Esteban Ortega Dominguez
• Mattia Varagnolo
• Sofia Pope Trogu

Timeline

2023-2024

Key Features

• Time Series Analysis: Implemented using fpp3, tsibble, and forecast packages.
• Data Visualization: Plots and graphical analysis are created with ggplot2 and plotly to explore trends in energy generation.
• Modeling: Includes models from lmtest, forecast, and prophet for various forecasting methods.
• Data Processing: Utilizes dplyr and lubridate to clean and structure the data for analysis.

Data Description

The project uses energy production data from various sources (e.g., Petroleum, Solar, Wind). Key steps include:

1. Converting date formats.
2. Summing generation by producer and energy source.
3. Removing missing or non-finite values for more accurate analysis.

Predictive Modeling

The project includes forecasting models for energy generation trends, particularly focusing on shifts from petroleum to renewable sources post-2010.

Libraries Used

• ggplot2, plotly: Data visualization.
• dplyr, lubridate: Data manipulation and time handling.
• forecast, prophet, fpp3: Time series analysis and forecasting.
• car: For Variance Inflation Factor (VIF) analysis.

Visualizations

The project includes time series plots for total energy generation by source over time, showcasing trends and forecasting future energy outputs.