Add a CLI that reads in an algorithm, loads data,
run the algorithm, and output performance metrics.
The examples are adapted to the new zipline API and
analyses are split into separate files.
Also add config files that run the example
algorithms with preset settings.
Highlights
- Reworked risk metrics, including verification against Excel spreadsheet
- Additional order methods
- Conversion of many data structures to use pandas
- Change to behavior of stop and limit orders (@pcawthron)
- Use pandas timezone handling throughout instead of Delorean
- New commission model (@stanh)
- Adds beginning of support for Toronto stock exchange. (@dstephens)
- Python 3 compatibility.
Unit tests now pass when run with Python 3.3, and Python 3 should
now be considered officially supported.
If anything does not work under Python 3, please file as a bug.
zipline.__version__ is now present. Closes#94.
Moreover, git master should have a .dev version string according
to convention. Releases then get the .dev label removed.
Highlights:
- Benchmark updating now permits empty ranges.
(Fixes runtime crash when running immediately after Easter 2013.)
- Risk metrics
- Performance improvents from converting to numpy and pandas.
<@wesm, wesmckinn@gmail.com>
- Refactoring of risk metric calculation out of class structure.
The pyandoc module throws an OSError of:
"No such file or directory",
when the underlying pandoc binary does not exist.
This error has caused confusion for numerous people during pip
installation.
pandoc is only needed on upload to PyPI to convert to ReST.
So, instead of doing a try/except on the `import pandoc` for all cases,
now checking whether or not 'upload' was invoked when calling setup.py,
so that only maintainers have to worry about installation of pandoc.
By only exercising the pandoc logic when
Highlights, with thanks to contributors inline:
- Runtime performance improvements
- Fixed the omission of peformance messages on days with no trades
- Changes to batch_transform implementation
-- supports sid filtering
-- performance improvements using pandas
-- added an option for only computating when there is a window length's
worth of data
- Added new risk metrics
-- Sortino
-- information ration
(Ryan Day, ryanday2@gmail.com @rday)
- Added stop and limit orders
(Tony Worm, verdverm@gmail.com @verdverm)
- Added variable recording
- Deprecated market_aware and delta kwargs to EventWindow
- Fixes to trading calendars for missing holidays
- Added TradingEnviorment context manager
- Added support for streaming through dividends
- Yahoo source now has OHLC
- Updates downloaded benchmark and treasury data when new data is available.
(Ryan Day, ryanday2@gmail.com @rday)
- Added optional adjustment of Yahoo data
(Jeremiah Lowin, jlowin@lowindata.com @jlowin)
Salient changes since last version:
- Adds non-holiday closings to trading calendar.
- Forward filling of missing treasury data.
- Improves handling of treasury data when the backtest's
end date day is not a market day.
- Adds option to forward fill data in batch transform.
Thomas Wiecki