This image implements the full Python scientific/mathematical/statistical stack (without complicated dependency hell), especially for use in financial economics, and includes modules to freely obtain the most current data (at no cost) in pandas format. Within seconds, the user can be fully engaged in creating original Jupyter notebooks using fecon235 as a learning template.
Keywords: Jupyter notebook pandas Federal Reserve FRED Ferbus Quandl GDP CPI PCE inflation unemployment wage income debt Case-Shiller housing asset portfolio equities SPX bonds TIPS rates currency FX euro EUR USD JPY yen XAU gold Brent WTI oil Holt-Winters time-series forecasting statistics econometrics
For complete details, see GitHub repository: https://github.com/rsvp/fecon235 which can also be found within the Docker container in the /opt/rsvp/fecon235 directory. Git itself is included, so pulling the latest commits is possible.
To RUN the pulled image:
$ docker run -p 8888:8888 -it rsvp/fecon235
which should yield a prompt, then execute:
and interact with your host browser at http://localhost:8888 which should begin by listing available Jupyter notebooks. By terminating nbstart, you will be returned to the container's Bash shell prompt, whereupon the IPython console can be called with modules for machine learning and econometrics.
Base image is derived from continuumio/anaconda
While our image tagged "latest" uses the most current development environment, the image rsvp/fecon235:v4.16.1030 uses rsvp/ana2-pd0181 for its base image, see https://hub.docker.com/r/rsvp/ana2-pd0181 for details. Note that the code is cross-platform, also python2 and python3 compatible.
Dockerfile source: https://git.io/fecon235-Dockerfile