Indicators/Strategies/Analyzers

• P

  Fisher Transform - Indicator
  • Praveen Baratam

  5
  5
  Posts
  138
  Views

  P

  I have written some code but it is not working. I believe I am missing something.

  import backtrader as bt import math __all__ = ['Fisher'] class Fisher(bt.Indicator): lines = ('h', 'l', 'scaled','Fx',) params = (('period', 20),) _alpha = 0.33 def __init__(self): self.addminperiod(self.params.period) self.l.h = bt.indicators.Highest(self.data.close, self.p.period) self.l.l = bt.indicators.Lowest(self.data.close, self.p.period) def next(self): d = self.data.close[0] h = self.l.h[0] l = self.l.l[0] scaled = self.l.scaled Fx = self.l.Fx scaled[0] = self._alpha * 2 * ((d - l) / (h - l) - .5) + (1-self._alpha) * scaled[1] if scaled[0] > 0.9999: scaled[0] = 1 elif scaled[0] < -0.9999: scaled[0] = -1 Fx[0] = .5 * math.log((1 + scaled[0]) / (1 - scaled[0])) + .5 * Fx[1]
• F

  MAE/MFE Analyzer
  • favetelinguis

  2
  2
  Posts
  74
  Views

  B

  Everything is possible. At the moment resources are low.

• C

  Choppiness Index (CHOP)
  • cgi1

  3
  3
  Posts
  81
  Views

  B

  Will be looked into ...

• W

  How i can run indicatirs.py on MT4
  • Watcharakit Haosaen

  2
  2
  Posts
  54
  Views

  B

  backtrader is an ecosystem and the indicators included in the package are tightly integrated in the platform. They are not TA-Lib like indicators/functions, which expose a call, take input buffers and parameters and return output buffers.

• J

  OLS example
  • jirathh

  2
  2
  Posts
  77
  Views

  J

  I tried to create a new indicator myself and try to get the last N days of data to do OLS and return the value to line, unfortunately my strategy get 0 in value.

  class toroOLS(bt.Indicator):
  #lines = ('slope','intercept',)
  lines=('slope',)
  params = (('period', 20),)

  def __init__(self): self.addminperiod(self.params.period) def next(self): closes = pd.Series(self.data.close.get(size=self.p.period)) x = np.arange(self.p.period).reshape(-1,1) x = sm.add_constant(x, prepend=True) #toro, intercept = sm.OLS(x, closes).fit().params #self.lines.slope[0] = toro #self.lines.intercept[0] = intercept toro = linear_model.LinearRegression().fit(x, closes) annualize_return = (((1+toro.coef_[0])**250) - 1)*100 self.lines.slope[0] = annualize_return

  ==========================================
  This is where the indicator is used. Can you please suggest?

  class TestStrategy(bt.Strategy):
  params = (
  ('exitbars', 5),
  )

  def log(self, txt, dt=None): ''' Logging function fot this strategy''' dt = dt or self.datas[0].datetime.date(0) print('%s, %s' % (dt.isoformat(), txt)) def __init__(self): # Keep a reference to the "close" line in the data[0] dataseries self.dataclose = self.datas[0].close self.slope = toroOLS(period=20)

  .....
  def next(self):
  # Simply log the closing price of the series from the reference
  self.log('Close, %.2f' % self.dataclose[0])
  self.log('Slope, %.10f' % self.slope[0])

• D

  Relative Data Input
  • Dixing Xu

  2
  2
  Posts
  76
  Views

  B

  @Dixing-Xu said in Relative Data Input:

  I wonder is there a way to easily convert close price to relative price based on the first bar of input. Or do I need to do it manually every time?

  You need a Filter which stores the 1st price seen during action and adjusts each price with regards to it. For reference see: Docs - Filters

  But such a filter actually has been implemented (it is used for some of the proprietary tools for the Backtrader Fund (see Esfera Capital - Funds

  It's a very simple filter

  class PercFilter(bt.metabase.ParamsBase): params = ( ('base', 100.0), ) def __init__(self, data): self._refclose = None def __call__(self, data, *args, **kwargs): if self._refclose is None: self._refclose = data.close[0] pc = 100.0 * (data.close[0] / self._refclose - 1.0) data.close[0] = self.p.base + pc return False # no change to stream structure/length

  @Dixing-Xu said in Relative Data Input:

  And will database such as sqlite be supported for storing data in the future?

  backtrader has no provisions for data collection/storage. Nothing prevents collecting data from the on-line sources, but how to actually store it is entirely up to the user. The sources contain a tool called rewrite-data which simply takes one of the inputs and outputs a csv in the expected default format for backtrader. The same data could be written to an sqlite database with a simple schema.

• R

  Datafeed as indicator
  • rm

  2
  2
  Posts
  94
  Views

  B

  @rm said in Datafeed as indicator:

  preprocessed data > original data

  That's a standard comparison and will generate a lines object. Same as if you did:

  data.close - data.high

  It doesn't matter if the operation is boolean or arithmentic.

• C

  Plot vertical line on price chart based on indicator value?
  • cemdev

  2
  2
  Posts
  79
  Views

  B

  Plotting a vertical line is actually not there, neither on the data nor on the indicator. Your best bets:

  Having an observer plotting markers on the data based on the value of an indicator. The BuySell observer looks into buy/sell orders to plot the markers. This would allow you to plot on the data but would be the most ad-hoc solution.

  Having an indicator plotting markers. For the markers you can look into the BuySell observer or the ParabolicSAR indicator. In your case and since you only want the marker when your indicator is 1, you only assign a value to the indicator when that's the case (and leave the 0 case untouched)

  From BuySell

  plotlines = dict( buy=dict(marker='^', markersize=8.0, color='lime', fillstyle='full', ls=''), sell=dict(marker='v', markersize=8.0, color='red', fillstyle='full', ls='') )

  The marker values and options are defined by matplotlib. See how this Blog - Arrows for the BuySell Observer changes the markers for the BuySell observer

• T

  3 Consecutive Highs / Low?
  • Taewoo Kim

  5
  5
  Posts
  198
  Views

  T

  @backtrader

  Will test it out. Thanks once again

• C

  Wrong value of Indicator in another CustomIndicator
  • cgi1

  6
  6
  Posts
  147
  Views

  C

  @backtrader said in Wrong value of Indicator in another CustomIndicator:

  Of course. You can pass any number of data-feed-like objects to indicators. They will be available as self.dataX or the corresponding self.datas[X]

  Nice. For reference:

  self.sr = SR(self.datas[0], self.atr)

  Gives

  inside custom indicator SR. So it can be used like

  class SR(bt.Indicator): def __init__(self, logger, master, fromdate, todate): # self.datas[0] is yahoo market data (as default) self.atr = self.datas[1]

  DONE

• 

  Heikin Ashi Delta indicator - Work but get warnings.
  indicator heikin ashi hadelta • • Paduel Gerion

  10
  10
  Posts
  432
  Views

  

  That I thought, Pythonized:

  upw = dtr==false and dtr[-1] and utr dnw = utr==false and dtr[-1] and dtr

  Meaning upw is true if current dtr is false and utr is true and the previous dtr is false. This would mark an upward trend change.

  The same

  upw = (not dtr) & dtr[-1] & utr dnw = (not dtr) & dtr[-1] & utr

  In Metastock:

  Assignation :=
  Comparison =

• G

  Make calcul with indicator
  • GwadaMan

  8
  8
  Posts
  203
  Views

  B

  @GwadaMan said in Make calcul with indicator:

  def __init__(self): self.sma0 = btind.MovAv.SMA(self.data, period=12) self.sma1 = ((self.sma0.lines.sma(0) - self.sma0.lines.sma(1)) / 2)

  A couple of comments for potential simplifications (some may for sure see it as more complex)

  self.sma0 = btind.MovAv.SMA(self.data, period=12)

  This can be usually simplified to:

  self.sma0 = btind.MovAv.SMA(period=12)

  When no specific data is specified, the 1st data of wherever the calculation is being made is implied.

  self.sma1 = ((self.sma0.lines.sma(0) - self.sma0.lines.sma(-1)) / 2)

  can be also be simplified to:

  self.sma1 = ((self.sma0.lines.sma - self.sma0.lines.sma(-1)) / 2)

  where there is no need to specify the (0), because the reference is implicit in the operation

  Although @ntguardian is right and (1) points to the future, this can be used for real reasons. For example to push values from the future into the past as it is done in the Ichimoku indicator. For example:

  ... self.l.chikou_span = self.data.close(self.p.chikou) ...

  This says that the current value of chikou_span has to be that of the future close (self.p.chikou has a default positive value of 26)

• C

  [FIXED] Ichimoku: Unable to get Future cloud
  • cgi1

  5
  5
  Posts
  153
  Views

  C

  @backtrader: Now works again as expected. Thanks!

• A

  Starting strategy later while using get(ago,size)
  • Andre

  5
  5
  Posts
  183
  Views

  B

  @Andre said in Starting strategy later while using get(ago,size):

  which calls Average(self.data, period=self.p.period). This seems to affect how the Strategy class operate

  No, it's not affecting how the Strategy operates. It simply informs the strategy about the minimum period contribution it has. See the following:

  Docs - Operating the Platfomr Docs - Strategy

  The Minimum Period concept is explained in both and what will be called when.

  The Average indicator is documented here:

  Docs - Indicator Reference (ordered alphabetically)

  If you want to read the code look in indicators/basicops.py

• V

  Arbitrage Strategy Test with Minute Data
  • vivalavida1982

  7
  7
  Posts
  356
  Views

  B

  A sincere recommendation again: read what you have before your eyes.

  First:

  For code/output blocks: Use ``` (aka backtick or grave accent) in a single line before and after the block. See: http://commonmark.org/help/

  Or else your posts are unreadable.

  Second:

  @vivalavida1982 said in Arbitrage Strategy Test with Minute Data:

  TypeError: unsupported operand type(s) for +: 'int' and 'str'

  You are obviously passing an str where an int is expected.

  Third.

  @vivalavida1982 said in Arbitrage Strategy Test with Minute Data:

  for i in self.data1['Date'] and self.data2['Date']:

  This seems to be a pandas-like syntax expectation. backtrader is not pandas. The expectation of your statement there is really a mistery here. Data feeds (like other lines objects) support integer indexing. See Docs - Platform Concepts

• 

  Hurst Exponent behavior in a Synthetic Trend Serie
  indicator hurst exponent sinthetic data • • Paduel Gerion

  7
  7
  Posts
  395
  Views

  

  These is a great solution. Thanks for taking into account my proposals and include them in this great framework.

• T

  Confused - How to close a long position vs. selling
  • Taewoo Kim

  2
  2
  Posts
  130
  Views

  B

  buy is for buying sell is for selling

  Both can obviously close an existing position if they are opposing each other. But if any of the above had intelligence to undo what the other do, how would you reverse a position?

  close is for closing a position and will take into account the existing position.

  They are documented in Docs - Strategy

  You may also close a position with order_target_xxx by setting the target to 0. See Docs - Target Orders

  And you can even close positions with sell if your position sizing is controlled with a Sizer or convert the same code to be a position reverser. See Docs - Sizers - Smart Staking

  The code will probably close positions in many cases, where there isn't simply a signal (neither buy nor sell)

• S

  Develop Heikinashi Indicators
  • santa3

  5
  5
  Posts
  227
  Views

  B

  This indicator is now in the development branch

• K

  Strategy with Signals
  • KT

  3
  3
  Posts
  172
  Views

  K

  Thanks @backtrader

• K

  Plot Indicator on Master Data
  • KT

  3
  3
  Posts
  150
  Views

  K

  Thanks @backtrader