Implement a writer for minute data into a format comprised of multiple
ctables, one for each individual asset, with a common 'index' shared by
all ctables where a given a dt maps to the same array index for all
equities and fields.
This format is pulled from the lazy-mainline/Q2.0 branch, with some
changes to the interface.
Add basic retrieval of values at a given dt to reader. Not yet used by
Zipline simulations, but added to support unit tests.
Also, rename stubbed out us_equity_minutes to minute_bars, since the
writer can be agnostic to asset type.
loaders
This allows people to set their cutoff time to the time they will
actually execute 'before_trading_start'. Currently this is just passed
to the constructor of the loader; however, I would like to make this
managed by the algorithm simulation runner. This would help keep all of
the loaders in sync and lock 'before_trading_start's execution to the
time the data is queried for.
In preparation for the incoming changes which no longer push every bar
through the tradesimulation, remove the adjustment of the period's cash on
every pricing change of a held futures asset.
Instead hold the last sale price for each held future either:
- At the end of each peformance period update the last sale prices of
all held futures, so that the pnl for the next period uses values
derived from the cash difference between the end of the two periods.
- When a transaction is processed for the Future, so that the correct
amount is applied to each cash adjustment. (i.e. the cash adjustment
is reset on every change of amount of the Future being held, so that
multiple size and prices do not need to be tracked for the same asset.)
Also, remove now unused dict of payout calculation modifier, since new
calculation reads the value directly off of the asset.
Remove update_last_sale test, since the method no longer returns a cash
value.
Instead of calling a function, where the only parameter is the tracker
object, make it a method, so that the snapshot of position tracker stats
can be more easily called as `pt.stats()`.
The test had a check for a pandas version (0.12.0) which was out of date
with the version in requirements, meaning the tests have not been run
regularly and unstable.
Skip via the decorator to make it more noticeable that tests are not
being run.
Refer to cumulative and todays performance explicitly instead of always
looping through.
The third value (minute) for which this was useful, has been removed.
Also, there are some actions where only cumulative may need application,
e.g. application of dividends. (However, this patch does not remove
dividend processing from todays performance, but opens up later patches
to make that distinction.)
EarningsCalendar loader.
- Moves most of AdjustedArray back into Python. The window iterator is
the only part that's performance-intensive.
- Adds a bootleg templating system for creating specialized versions of
AdjustedArrayWindow for each concrete type we care about.
- Adds support for differently dtyped terms in pipeline. This allows us
to use datetime64s which are needed in the EarningsCalendar.
- Adds EarningsCalendar dataset for the next and previous earnings
announcements in pipeline.
- Adds in memory loader for EarningsCalendar.
- Adds blaze loader for EarningsCalendar.
Return -1 when there is a zero value for a spot price.
Intended for use by the incoming data portal changes. When the data
portal will see a -1 value, the portal will seek back a trading day
until a non-negative value is returned.
Volumes were incorrectly having the thousands factor applied, however
the volume is written as is (without the factor, since it volume is an
int, not float value.)
Fix by adding a special case for volume which returns the price as is.
llllllllll