Survey of fuel prices in Brazil: Geospatial Analysis

This project includes EDA and geospatial analyses in the form of choropleth maps to visualize fuel price data from the Agência Nacional do Petróleo, Gás Natural e Biocombustíveis in Brazil. The dataset was made available through Kaggle and incorporates public geospatial data available through public ftp from the Brazilian government.

— Project Status: [Completed]

Project Objective

The purpose of the project is to perform basic exploratory data analysis and to answer three questions regarding the prices of fuel products in Brazil using geospatial analysis:

• How did the price change for the different regions of Brazil?
• Within a region, which states increased their prices the most?
• Which states are the cheapest (or most expensive) for different types of fuels?

These questions are addressed using choropleth map visualizations in the notebook 0.3-geospatial-analysis.ipynb.

Description

The fuel price data comes from the Agência Nacional do Petróleo, Gás Natural e Biocombustíveis (ANP in Portuguese), which releases weekly price reports of gas/petrol, diesel and other fuels used in transportation across the country. It includes the mean value per liter, number of gas stations analyzed and other information grouped by region and state. The analysis is enriched by joining the price data with geospatial data from the Brazilian government.

Technologies

• Python, Anaconda
• Pandas, Matplotlib, Seaborn, Geopandas, Jupyter Notebook

Data Sources

Fuel Prices:

• Aggregated dataset: https://www.kaggle.com/matheusfreitag/gas-prices-in-brazil
• Original data source: http://www.anp.gov.br/precos-e-defesa-da-concorrencia/precos/levantamento-de-precos/serie-historica-do-levantamento-de-precos-e-de-margens-de-comercializacao-de-combustiveis

Shapefiles:

• ftp://geoftp.ibge.gov.br/organizacao_do_territorio/malhas_territoriais/malhas_municipais/municipio_2016/Brasil/BR/

Featured Notebooks

• 0.1-import-clean-eda.ipynb —> view notebook
• 0.2-geospatial-analysis.ipynb —> view notebook
• An older version of this project was submitted as a Kaggle kernel.

Getting Started

1. Clone repo
2. Run conda env create -f environment.yml to create the environment (requires Anaconda)
3. Run conda activate brazil_fuel_price_env to activate the environment
4. Run each notebook:
   • 0.1-import-clean-eda.ipynb
   • 0.2-geospatial-analysis.ipynb
5. Figures are saved to the figures directory

Authors

• Gregory Lindsey - abk7777

License

This project is licensed under the MIT License.