CoffeClimateCorrelation

In this project, we analyzed whether coffee sales in Tübingen are correlated to weather variables such as temperature, sunshine duration and precipitation.

Datasets

For our weather dataset, we chose the historic weather data recordings from ECAD (European Climate Assessment & Dataset). We chose the weather station in Stuttgart-Echterdingen. The data is available from 1953 to 2022. The data is available in the folder /raw_weatherdata.

For our coffee sales dataset, we got access to the sales data of a local café, the SUEDHANG. The data is available from 2020 to 2022. The data is located in the folder /raw_coffeesalesdata. However, raw coffee sales data is available only on request due to privacy reasons, here only the processed data is available in the folder /processed_data.

Data processing

The data processing is done in the jupyter notebooks. The data processing is done in the following order:

1. 01_find_coffeproducts.ipynb (Does not work without access to the raw coffee sales data)
2. 02_load_coffeesalesdata.ipynb (Does not work without access to the raw coffee sales data)
3. 03_load_weatherdata.ipynb (Does work without access to the raw weather data)

Data analysis and results

The data analysis is done in the jupyter notebook 04_plot_combined_data.ipynb. Plots are saved in the folder /results.

Authors

Nadia Vohwinkel and Magnus Kaut