260 views
# Initial investigations: Hourly Data for Zürich I started investigating the hourly dataset of the city of Zürich (Found [here](https://data.stadt-zuerich.ch/dataset?q=luftqualit%C3%A4t)). ## Data preparation So far, the most time was spent in data tidying and cleanup. The main problem is that the file has multiple headers, namely both the station name and the measurement description. With the use of tidyr I cleaned up the data and produced a [tidy](https://cran.r-project.org/web/packages/tidyr/vignettes/tidy-data.html) dataset. Where date, measurement values and station names are stored in separate columns. ## Exploratory analysis As discussed over the chat, we focused on the nitrogen dioxide measurements and decided not to pursue the study of ozone and PM10. Here I plotted the timeseries of nitrogen dioxide over time. We can see an increasing trend, but we must be aware of confounding variables before coming to any conclusion. ![](https://us-east-1.linodeobjects.com/dribdat/uploads/upload_42088150050741666156b7e8f4e71e8d.png)