Hedging algorithms with lag
[![Build Status](https://img.shields.io/travis/reichlab/ledge.svg?style=flat-square)](https://travis-ci.org/reichlab/ledge) [![Pypi](https://img.shields.io/pypi/v/ledge.svg?style=flat-square)](https://pypi.python.org/pypi/ledge)
Hedging algorithms with lag.
Besides the regular setting of ‘prediction with expert advice’, we are interested in working with truth values with _lags_. This results in a partially observed truth value for the present (and the past) at each step. In a discrete time setting with delays in accurate characterization of _final_ truth, a truth value is specified by:
- timepoint: The time itself
- lag: Time the value was revealed - timepoint
ledge is a composed from a bunch of types and functions. It works with [DataArrays](http://xarray.pydata.org/en/stable/generated/xarray.DataArray.html#xarray.DataArray) for model _predictions_, _losses_ and _truths_.
There is no single model and the user is supposed to _compose_ a model using the components in here. Each module is written as a literate [org-mode](https://orgmode.org/) file and contains usage documentation for the containing functions. Start with the main file ./ledge/README.org.
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
|Filename, size||File type||Python version||Upload date||Hashes|
|Filename, size ledge-0.0.7.tar.gz (17.3 kB)||File type Source||Python version None||Upload date||Hashes View|
|Filename, size ledge-0.0.7-py3-none-any.whl (7.4 kB)||File type Wheel||Python version py3||Upload date||Hashes View|