Running the Jupyter stock trading notebooks in Azure Notebooks

Update October 2020:  Azure Notebooks Preview will be retired from January 15th, 2021 and all user data will be deleted - for alternatives take a look at https://docs.microsoft.com/en-us/azure/notebooks/quickstart-export-jupyter-notebook-project

In the previous blog posts - Using Euronext stock data from Quandl in Jupyter notebooks and Working with multiple time series trading data from Quandl in Jupyter Notebooks I showed how you can use Jupyter notebooks to analyze stock trading running the Anaconda distribution from my local machine. The notebooks are available on Github - https://github.com/jorisp/tradingnotebooks

It is however also possible to run them in Microsoft Cloud using Azure Notebooks on  https://notebooks.azure.com/ . Azure Notebooks is a free hosted service  to develop and run Jupyter notebooks in the cloud along with supporting packages for Python, R and F#. You can just login and get started without having to setup or install anything and run the code within your browser.

On the home page you will see a number of Jupyter notebook projects which you can clone into your own personal library.

By default, projects run on the Free Compute tier, which is limited to 4GB of memory and 1GB of data. You can bypass these limitations and increase compute power by using a different virtual machine that you've provisioned in an Azure subscription. For more information see Use Azure Data Science Virtual Machines

To be able to run the Jupyter notebook from https://github.com/jorisp/tradingnotebooks/blob/master/Quandl_API_Euronext_ABI_Shared.ipynb on Azure notebooks, the only thing you need to do is installing the Quandl Python module by adding !pip install quandl in a new code cell on the top.

The free compute tier from Azure notebooks is running a custom version of Anaconda but not always the latest version - if you need specific Python modules you need to be aware that the installed version might not always be the latest version