I worked for several years with weather data as a potential moderator in different marketing models. We analyzed and predict the impact of different weather variables on sales and promotion effectiveness. One challenge during these projects always was to aggregate the weather data in a meaningful manner to use them in our models. While meteorologist often work on hourly grid data, our data in the retailing context was by best daily and state level.
Within the last year I was contacted by colleague which data sources are available, how to aggregate the data and which variables are meaningful. Therefore, I decided to use the public available data of the German Weather Service and aggregate there to the level we often need.
You can download the data as csv-file and cite them as follows.
If you have any other request, different time span or different aggregation level, do not hesitate to contact me [wiebke.keller@uni-tuebingen.de].
Daily Weather Data Germany
Keller, Wiebke I. Y., & Keller, Jan D. (2021): „Daily weather data averages for Germany aggregated over official weather stations“ [Data set Zenodo].
http://doi.org/10.5281/zenodo.5052777
Description of the data:
Daily data for several weather variables. The data is averaged across German weather stations and day for the years 2016 till 2020:
- temperature_mean:
mean daily temperature in degree Celsius averaged across all weather stations in Germany. - temperature_max:
maximum daily temperature in degree Celsius averaged across all weather stations in Germany.
- precipitation:
daily precipitation sum in millimeter (equals liter per square meter) averaged across all weather stations in Germany.
- sunshine:
sunshine duration per day averaged across all weather stations in Germany.