@run-out said in How to extract historical account value from backtrader or its results:
Dictionary
dictionary = results[0].analyzers.getbyname("cash_market").get_analysis() df = pd.DataFrame(dictionary).T df.columns = ["Date", "Cash", "Value"]
It works, thank you so much.
I guess the system double_unders can't be reached from outside of the object
I have a trading bot using Oanda that I'm trying to get working. Oanda for the pair that I'm considering using 50x leverage, and I want to reduce this to 5. The easiest way that I've though of is to reduce the cash available to the system to the actual trading results achievable on oanda will be simulated by backtrader. Essentially I want to remove cash from the broker and add it to the fund, making a backup account that is used to adjust the size of the actual account I'm using for trading calculations.
As the first step to implementing this, I use the following code:
def nextstart(self):
des_leverage = 5
act_leverage = 50
starting_cash = self.broker.get_cash()
use_cash = starting_cash * des_leverage / act_leverage
adjustment = use_cash - starting_cash
self.broker.add_cash(adjustment)
current_cash = self.broker.get_cash()
print('starting cash:$ {}' . format(starting_cash))
print('desired cash due to leverage:$ {}' . format(use_cash))
print('adjustment: $ {}' . format(adjustment))
print('current cash: $ {}' . format(current_cash))
The results are below. I haven't implemented the adding the subtracted money to the fund, but it doesn't look like my add_cash is working. Any ideas?
starting cash:$ 1000.0
desired cash due to leverage:$ 100.0
adjustment: $ -900.0
current cash: $ 1000.0
Thanks.
Am I correct in assuming that backtrader doesn't exit trades when a margin call would have happened? I have a system I'm testing in forex which can be highly leveraged but I want to use a lesser amount. I've set the leverage to 4 (from 50 that is available at many forex outlets) but I'm not able to see that any trades have been exited for a margin exceedance if the market goes against the trade and the account value falls below the required amount for leverage.
It appears that new trades are rejected due to margin if there is insufficient money in the trading account, but I don't see where existing trades are exited due to margin.
I haven't seen this explicitly stated in the documentation yet but I might have missed it.
Thanks.
@run-out said in How to extract historical account value from backtrader or its results:
Dictionary
dictionary = results[0].analyzers.getbyname("cash_market").get_analysis() df = pd.DataFrame(dictionary).T df.columns = ["Date", "Cash", "Value"]
It works, thank you so much.
I guess the system double_unders can't be reached from outside of the object
I know this should be fairly easy to do but I'm stumped. I want to get the account value by bar from a cerebro instance so I can do analysis on a spreadsheet.
For some reason I can't seem to figure out how to make this happen. I have this observer:
class AcctValue(bt.Observer):
alias = ('Value',)
lines = ('value',)
plotinfo = {"plot": True, "subplot": True}
def next(self):
self.lines.value[0] = self._owner.broker.getvalue() # Get today's account value (cash + stocks)
which I call with this code block:
# Create an instance of cerebro
cerebro = bt.Cerebro()
# Input a starting broker account value
cerebro.broker.setcash(startcash)
# Add our strategy
cerebro.addstrategy(Boll)
comminfo = forexSpreadCommisionScheme(spread = spread, leverage = leverage)
comminfo.show_spread_leverage()
cerebro.broker.addcommissioninfo(comminfo)
# Add the data to Cerebro
cerebro.adddata(data)
# Add a sizer
cerebro.addsizer(ForexSizer,risk_per_trade=.0025,ATR_mult=6)
#cerebro.addobserver(OrderObserver)
cerebro.addobserver(AcctValue)
cerebro.addobservermulti(bt.observers.BuySell, markersize=4.0) # Plots up/down arrows
# Add the analyzers we are interested in
#cerebro.addanalyzer(bt.analyzers.GrossLeverage, _name="gl")
cerebro.addanalyzer(bt.analyzers.TradeAnalyzer, _name="ta")
cerebro.addanalyzer(bt.analyzers.SQN, _name="sqn")
cerebro.addanalyzer(bt.analyzers.Returns, _name="ret")
cerebro.addanalyzer(bt.analyzers.SharpeRatio, _name='mysharpe')
cerebro.addanalyzer(bt.analyzers.AnnualReturn, _name='ann')
cerebro.addanalyzer(bt.analyzers.DrawDown, _name='dd')
cerebro.addanalyzer(bt.analyzers.Transactions, _name = 'transactions')
#cerebro.addanalyzer(AcctStats)
# Run over everything
strategies = cerebro.run()
firstStrat = strategies[0]
result=strategies[0]
#Get final portfolio Value
portvalue = cerebro.broker.getvalue()
pnl = portvalue - startcash
#Print out the final result
print('Final Portfolio Value: ${}'.format(round(portvalue,2)))
print('P/L: ${}'.format(round(pnl,2)))
printSharpe(firstStrat.analyzers.mysharpe.get_analysis())
printTradeAnalysis(firstStrat.analyzers.ta.get_analysis())
printSQN(firstStrat.analyzers.sqn.get_analysis())
printAnnualReturns(result.analyzers.ann.get_analysis())
printReturns(result.analyzers.ret.get_analysis())
printDD(result.analyzers.dd.get_analysis())
using
b= cerebro.Observer.AccountValue()
b
doesn't work.
When I look at the vars(result) I see:
<main.AcctValue at 0x7f931b97a610>
but I don't know how to access this.
Can anyone help me?
I tried that solution as well as various permutations of the docs. It seems like there is something different from the way bactrader works vs the btoandav20 works.
I may just have to roll my own solutions for some of these, I'm not sure. I'm able to stream the data, enter new orders, but the cancel stuff I can't seem to do.
I'm trying to code something to test with OANDA. I want to ensure that any existing orders are closed before I begin.
I'm working with btoandav20. I can't seem to find an example of canceling an order and my guesses aren't getting me anywhere.
Thanks. Code follows.
import backtrader as bt
import btoandav20 as bto
import json
import datetime
import math
''' Order info '''
class TestStrategy(bt.Strategy):
params = (
("period", 20),
("devfactor", 2.5),
("atr_length", 50),
#("size", 9000),
("debug", False),
("verbose", False),
('atr_stop', 1),
('atr_profit', 2)
)
def __init__(self):
self.orders = None
self.prev_size = 0
#Provides any notificaitons about the data.
def notify_data(self, data, status, *args, **kwargs):
#print('*' * 5, 'DATA NOTIF:', data._getstatusname(status), *args)
pass
def notify_store(self, msg, *args, **kwargs):
#print('*' * 5, 'STORE NOTIF:', msg)
pass
def notify_order(self, order):
if order.status in [order.Completed, order.Cancelled, order.Rejected]:
self.order = None
print('-' * 50, 'ORDER BEGIN', datetime.datetime.now())
print(order)
print('-' * 50, 'ORDER END')
def notify_trade(self, trade):
print('-' * 50, 'TRADE BEGIN', datetime.datetime.now())
print(trade)
print('-' * 50, 'TRADE END')
def next(self):
print('trying to cancel orders')
#orders = self.orders()
#print(orders)
for o in self.broker.orders:
print("order info: {} " . format(o))
self.broker.cancel(o)
bt.Order.cancel(o)
#for o in bt.broker.orders:
# print('order info: {} ' .format(o))
txt = list()
#txt.append('Data0')
#txt.append('%04d' % len(self.data0))
dtfmt = '%Y-%m-%dT%H:%M:%S.%f'
#txt.append('{:f}'.format(self.data.datetime[0]))
txt.append('%s' % self.data.datetime.datetime(0).strftime(dtfmt))
txt.append('{:f}'.format(self.data.open[0]))
txt.append('{:f}'.format(self.data.high[0]))
txt.append('{:f}'.format(self.data.low[0]))
txt.append('{:f}'.format(self.data.close[0]))
txt.append('{:6d}'.format(int(self.data.volume[0])))
txt.append('{:d}'.format(int(self.data.openinterest[0])))
#txt.append('{:f}'.format(self.atr[0]))
print(', '.join(txt))
if len(self.datas) > 1 and len(self.data1):
txt = list()
txt.append('Data1')
txt.append('%04d' % len(self.data1))
dtfmt = '%Y-%m-%dT%H:%M:%S.%f'
txt.append('{}'.format(self.data1.datetime[0]))
txt.append('%s' % self.data1.datetime.datetime(0).strftime(dtfmt))
txt.append('{}'.format(self.data1.open[0]))
txt.append('{}'.format(self.data1.high[0]))
txt.append('{}'.format(self.data1.low[0]))
txt.append('{}'.format(self.data1.close[0]))
txt.append('{}'.format(self.data1.volume[0]))
txt.append('{}'.format(self.data1.openinterest[0]))
txt.append('{}'.format(float('NaN')))
print(', '.join(txt))
def start(self):
#if self.order:
# self.order.cancel()
if self.data0.contractdetails is not None:
print('-- Contract Details:')
print(self.data0.contractdetails)
print('Started')
acc_cash = cerebro.broker.getcash()
acc_val = cerebro.broker.getvalue()
print('Account Cash = {}'.format(acc_cash))
print('Account Value = {}'.format(acc_val))
#orders = self.broker.get_orders_open()
#if:
# self.cancel(self.order)
#if orders:
# for order in orders:
# self.broker.cancel(order)
with open("config.json", "r") as file:
config = json.load(file)
storekwargs = dict(
token=config["oanda"]["token"],
account=config["oanda"]["account"],
practice=config["oanda"]["practice"],
#use_positions = True,
#notif_transactions = True,
stream_timeout=2,
account_poll_freq = 1
)
store = bto.stores.OandaV20Store(**storekwargs)
datakwargs = dict(
timeframe=bt.TimeFrame.Minutes,
compression=30,
#tz='Europe/Berlin',
tz = 'UTC',
bidask = False,
backfill=False,
backfill_start=False,
)
data = store.getdata(dataname="EUR_USD", **datakwargs)
#data.resample(
# timeframe=bt.TimeFrame.Minutes,
# compression=30) # rightedge=True, boundoff=1), compression is the minutes or seconds
cerebro = bt.Cerebro()
cerebro.adddata(data)
#broker = bto.brokers.OandaV20Broker()
cerebro.setbroker(store.getbroker())
cerebro.addstrategy(TestStrategy)
cerebro.run()
I feel like an idiot; I finally figured out that I wasn't passing data to the bt.indicators.ADX and ATR statements. Apparently the system uses the first data for any subsequent calls if you don't send it new data.
self.adx[dat] = bt.indicators.ADX(dat, period = self.params.n, plot=True)
self.atr[dat] = bt.indicators.ATR(dat, period = self.params.n, plot=True)
Is the correct formulation of those two lines.
I saw this in an old post and am trying to get it to work correctly.
I'm trying to generate ADX and ATR values for my symbols. I create a list with the symbols in it.
I don't have the whole thing built; I was trying to build step-by-step but I must be doing something wrong as the ADX and ATR seems to be calculated for the first symbol in the list . The data shows that it is different in the graph but the indicators don't for some reason.
Can anyone point me in the right direction?
import backtrader as bt
import pandas as pd
import barchart_data as bd
class ADX_C(bt.Strategy):
params = {"n": 25}
def __init__(self):
"""initialize the strategy"""
self.adx = dict()
self.atr = dict()
for dat in self.datas:
self.adx[dat] = bt.indicators.ADX(period = self.params.n, plot=True)
self.atr[dat] = bt.indicators.ATR(period = self.params.n, plot=True)
symbols=["ZROZ","XMPT"]
min_per_bar=1440
begin_date = '2019-01-01'
end_date = '2020-06-11'
cerebro = bt.Cerebro()
for s in symbols:
data = bd.BarchartData(s,min_per_bar).data_by_dates(begin_date,end_date)
bt_data = bt.feeds.PandasData(dataname=data, name = s)
cerebro.adddata(bt_data, name = s)
cerebro.addstrategy(ADX_C)
cerebro.run()
cerebro.plot(iplot=True, volume=False, width=50, height=30, dpi=600)
The following is my attempt to make a custom position sizer based on ATR and the multiplier for the given future. I'm sure this is relatively easy but I seem to have hit a wall.
I seem to have a position size of 0 for every trade.
What am I getting wrong?
Thanks.
class comm_sizer(bt.Sizer):
def __init__(self, atr):
#self.atr_parameter= atr_parameter
atr= self.atr
risk_per_trade=5000
atr_risk_per_trade=5
sizing = math.floor((risk_per_trade/(mult*atr_risk_per_trade*atr)))
# position = self.broker.getposition(data)
def _getsizing(self,comminfo,cash,data,isbuy):
return sizing
I have a system where I enter a long position and simultaneously enter a protective stop. I have a separate exit that sometimes occurs. When this occurs, I'd like to be able to cancel the protective stop. I currently use more than one instrument, (currently looking at a basket of 5 stocks), so I'd like to either find a way to determine which orders are still open by symbol or cancel all open orders by symbol.
After looking through the documentation I believe this should be possible but I can't seem to get it to work.
Does anyone have any ideas?
Thanks.