1.9.60.122

• Remove unused files
• README update, Docstring corrections, documentation corrections
• Update travis settings

Available via pip

Thank you for the message. All options are being considered (messages from other threads were also conveyed)

1.9.66.122

• Fix regression introduced with 8f537a1c2c271eb5cfc592b373697732597d26d6
  which voids the count of lost trades
• Allow rollover to distinguish between no values temporarily (with None)
  and no values permanently (with False)
• Avoid math domain error for negative returns in logarithmic calculations
• Fix local variable declaration for compound returns
• Fix typo in date2num tz conversion which shows up in direct usage

The x-axis is there (luckily, because if not everything would be displayed on a singularity rim and we should fear a potential earth implosion into such an event) but the x-ticks (i.e: timestamps in this case) are not displayed in the latest latest version of matplotlib (currently 3.1.1)

The previous stable version 3.0.3 and the last LTS version 2.2.4 do display the x-ticks.

pip install --force matplotlib==YOUR-PREFERRED-WORKING-VERSION

Let's try to summarize:

• Non-Live Data feeds have well defined interfaces and it is documented

  You only have to override _load(self), which will be in charge of loading the values into the lines of the data series (in most cases these will be: datetime, open, high, low, close, volume and for futures openinterest

  It returns True if it has been able to fill the next set of values or False to indicate the end of the data stream.

  What's left for the implementer: the decision as to how _load(self) receives the values from the feed. Examples:

  • From a file: for each call to _load you can simply read the next line, process the data, fill the lines and return, until EOF is met.

    There is even documentation of how to do it for a binary file Docs - Binary Datafeed Development

    This has been for example generalized for CSV-based sources by adding a _loadline method, which receives the line broken down into tokens. In this case only overriding of _loadline is needed. See Docs - CSV Data Feed Development

• Live Data Feeds could have a couple of things added

  • Methods islive and tzoffset, the chosen pattern for data reception and the extra return value for _load which is None to indicate that the data feed has currently nothing to deliver but could have it later (the data stream is active but has not come to an end)

    The problem with the rest ... it is provider dependent. And the hackhaton (aka BacktraderCon 2017) last weekend has proven it. The initial implementation the guys had followed the guidelines from the Oanda implementation, but because the provider (Kraken) only offers polling and has low limits for its Rate Limiting policy, everything is a historical download at the end of the day. Suddenly instead of 2 queues, both queues are the same but the usage is different.

• Broker have a well defined interface

  Here it is really a matter of work in the store, which offers a private interface to the broker. Through this interface for example the store will convert a order from backtrader to an order which is palatable by the real broker. On the way back, broker notifications have to be adapted, to change the status of orders appropriately.

A kind of paper to give guidelines can be considered (will be done so to say), but at the end of the day and the guys from the BacktraderCon could tell a lot about it, it's about the very small details from each broker.

A summary of what's on the table:

Time Scheduled Functions/Callbacks [Done]

• Functions will be called at specific times (periodical calling like ... "every 60 minutes starting at 08:30" may be considered)

• Timezone considerations

  • The strategy shows the time UTC-like format because it it not tz bounded like the data feeds
  • pytz doesn't have a notion of local timezone. This poses a small challenged when it comes to accept times which are referring to the actual local time (it may not seem so in real time, but it is in backtesting, in which the datetime reference is marked by the data feeds being backtested)
  • And even if local time is available, it may not be the wished reference, because the traded data feed can be in a different timezone (some of the use cases show 3 timezones in play: user, data 1 and data 2)

• The best possible place to add the scheduling seems to be in the strategy's __init__ method and as such should be a method of Strategy

• This doesn't preclude having a global scheduling which could be entered in cerebro to control any strategy.

Collection of numpy/pandas/statsmodel dependent indicators [Done]

• The discussions here have given birth to some.

• So far none has been added to the package for a good reason: backtrader was meant to be a pure Python package, in the sense that it wouldn't need any external dependencies beyond a regular Python distribution.

  (With the exception of matplotlib if plotting was wished)

• But at some point in time and even it not based on pandas or numpy arrays, those indicators should make it into the main distribution

Reverting Resampling/Replaying to the original model [Discarded]

• The original mode was to create another data feed which would then deliver own bars
• At some point in time both features were rewritten as filters manipulating the data stream, which has proven to be prone to quirks, problems and complicated code
• It seems sensible to rewrite the functionalities again as originally envisioned

Maybe

• Genetic Optimization

v2

• Once the above items are done ... v2 can be kickstarted to try to

  • Apply the same general ideas with regards to ease of use
  • Increase speed

This is a very recent error introduced with this commit: https://github.com/backtrader/backtrader/commit/8f537a1c2c271eb5cfc592b373697732597d26d6

In attempting to fix the bool problem if you only had 1 trade the proper not to identify lost trades was lost.

This is now fixed in this push to the development branch: https://github.com/backtrader/backtrader/commit/cc2751a5f53166f68c5340eb876579f1a5590bf5

The event is over ...

• We watched the Champions League Final
• We ate Bavarian food

and we even ...

• Did some backtrading

A repository with the snippets:

• GitHub - BacktraderCon

There you will also find a PDF with the notes that were taking during the event (including the pictures)

• GitHub - BacktraderCon 2017 Notes

And of course ... the best picture ever

Some of the contributions which were in the queue would for sure be nice additions for some users of `backtrader´. In order to allow for a seamless integration of those for as many people as possible, the following approach will be taken

• A side project named backtrader_contrib will be created

• This project will accept pull requests

• Elements submitted (indicators, analyzers, ...) will be injected into the main backtrader package

  • If a component generates an exception during importing it won't be injected

• The licensing is left open to the user with one exception: no GPLv2. Being the reason that some licenses like Apache License 2.0 is not compatible with GPLv2, but it is with GPLv3

  See: GNU - A Quick Guide to GPLv3 and GNU - Various Licenses and Comments about Them

  The repository will contain the license text for GPLv3, ASF 2.0, MIT, BSD

• The repository will be published as a pip package on a regular basis

You read it right and has been corrected. The platform was tightened around that area due to how users were actually addressing the multiple possible corners.

Trade documentation is here: [Docs - Trade][https://backtrader.com/docu/trade/]

Member Attributes:

• ref: unique trade identifier

• status (int) : one of Created, Open, Closed

• tradeid: grouping tradeid passed to orders during creation
  The default in orders is 0

• size (int) : current size of the trade

• price (float) : current price of the trade

• value (float) : current value of the trade

• commission (float) : current accumulated commission

• pnl (float) : current profit and loss of the trade (gross pnl)

• pnlcomm (float) : current profit and loss of the trade minus
  commission (net pnl)

• isclosed (bool) : records if the last update closed (set size to
  null the trade

• isopen (bool) : records if any update has opened the trade

• justopened (bool): if the trade was just opened

• baropen (int): bar in which this trade was opened

• dtopen (float): float coded datetime in which the trade was
  opened

  • Use method open_datetime to get a Python datetime.datetime
    or use the platform provided num2date method

• barclose (int) : bar in which this trade was closed

• dtclose (float) : float coded datetime in which the trade was
  closed

  • Use method close_datetime to get a Python datetime.datetime
    or use the platform provided num2date method

• barlen (int ) : number of bars this trade was open

• historyon (bool) : whether history has to be recorded

• history (list) : holds a list updated with each “update” event
  containing the resulting status and parameters used in the update

1.9.78.123

• PR#479 Fix errors for simulated orders

1.9.77.123:

• PR#472
  • Added posibitity for Black theme for charts
  • Added posibitity to run on matplotlib 3.6+
  • Added posibitity to run on Python 3.9+

1.9.76.123:

• PR#405 Fix initial Renko bricks
• Add option to select fixing of initial Renko bricks
• PR#403 partial order execution iterpending reported incorrectly
• PR#402 bug fix: #5 fixing writer.py after 1.9.75.123 pull
• PR#406 trade.py upgraded to be able to be unpickled. (#406)
• PR#411 [bug fix] frompackages directive functionality seems to be
  broken when using inheritance (#411)
• Typo corrections PR#409, PR#407

Version 1.9.78.123 is pushed

There is a post which shows monthly rebalancing.

• Here: https://www.backtrader.com/blog/2019-07-19-rebalancing-conservative/rebalancing-conservative/

Strategies using 1/3 of the capital

• The easiest way is to divide it at the time of the creation of cerebro and adding each Strategy
• If not, strategies that are winning have to rebalance, using a maximum of the total net asset value, to free capital for the losing strategies. To avoid the inherent problem of sequential evaluation and that losing strategies could be looked at before, the sell orders have to be executed on a given day and the buy orders (with freed capital from the selling) on the next day.
• The previous thing adds a problem which is that the data cannot be used with a monthly period because the next day would be the next evaluated bar, i.e.: the next month actually.

Evaluating at the end of the month

• Given that the gregorian calendar has months of different length, the best way is to use the first day seen of a month (which must not be 1 if it's a weekend or a bank holiday) and execute the rebalancing.

• You simply need to keep a sentinel value keeping the last month (or None at the start) in which rebalancing was done and act when the month changes

Different starting dates and 3-month period

• Because you have 3 different dates and they are set 1 month apart, this is actually granted, because you won't receive data in your next until the 3 strategies deliver data.

• The 1st time you will see data is on 2013-03-01 and you can use 3 as the sentinel for the last execution

• When the date changes to April and with it to 4, you may execute the 1st rebalancing.

See, I do understand that you have a problem but,

1. Pictures cannot be used to reproduce problems
2. Even if one typed in the entire text of your picture:
   • It is not a complete script
   • It contains no sample data

To start with so that you may grasp the difficulties of attempting something:

• What is res?

In any case I do stand by my early diagnostic. The data feed has not been started, because assuming I believe the return of res is a dataframe, that's not how to add it to cerebro

See the documentation for PandasData: https://www.backtrader.com/docu/pandas-datafeed/pandas-datafeed/

Pull-request has been merged. Version 1.9.77.123 has been generated

@guslijif

Quite difficult to reproduce anything from a picture. We don't even know how that's loaded, hence the comment before about a minimum reproducible sample.

It's going to be looked upon, this week (CW15 2023) unless the sky happens to fall on our heads.