def calculate_signals(self, event):
if (event.type == EventType.BAR and event.ticker == self.ticker):
# Add latest adjusted closing price to the
# short and long window bars
self.lw_bars.append(event.adj_close_price)
if self.bars > self.long_window - self.short_window:
self.sw_bars.append(event.adj_close_price)
# Enough bars are present for trading
if self.bars > self.long_window:
# Calculate the simple moving averages
short_sma = np.mean(self.sw_bars)
long_sma = np.mean(self.lw_bars)
# Trading signals based on moving average cross
if short_sma > long_sma and not self.invested:
print("LONG %s: %s" % (self.ticker, event.time))
signal = SignalEvent(self.ticker,
"BOT",
suggested_quantity=self.base_quantity)
self.events_queue.put(signal)
self.invested = True
elif short_sma < long_sma and self.invested:
print("SHORT %s: %s" % (self.ticker, event.time))
signal = SignalEvent(self.ticker,
"SLD",
suggested_quantity=self.base_quantity)
self.events_queue.put(signal)
self.invested = False
self.bars += 1
def calculate_signals(self, event):
# Applies SMA to first ticker
ticker = self.tickers[0]
if event.type == EventType.BAR and event.ticker == ticker:
# Add latest adjusted closing price to the
# short and long window bars
price = event.adj_close_price / float(PriceParser.PRICE_MULTIPLIER)
self.lw_bars.append(price)
if self.bars > self.long_window - self.short_window:
self.sw_bars.append(price)
# Enough bars are present for trading
if self.bars > self.long_window:
# Calculate the simple moving averages
short_sma = np.mean(self.sw_bars)
long_sma = np.mean(self.lw_bars)
# Trading signals based on moving average cross
if short_sma > long_sma and not self.invested:
print("LONG: %s" % event.time)
signal = SignalEvent(ticker, "BOT", self.base_quantity)
self.events_queue.put(signal)
self.invested = True
elif short_sma < long_sma and self.invested:
print("SHORT: %s" % event.time)
signal = SignalEvent(ticker, "SLD", self.base_quantity)
self.events_queue.put(signal)
self.invested = False
self.bars += 1
def calculate_signals(self, event):
if (event.type in [EventType.BAR, EventType.TICK]
and self._end_of_month(event.time)):
ticker = event.ticker
if self.tickers_invested[ticker]:
liquidate_signal = SignalEvent(ticker, "EXIT")
self.evert_queue.put(liquidate_signal)
long_signal = SignalEvent(ticker, "BOT")
self.evert_queue.put(long_signal)
self.tickers_invested[ticker] = True
def go_short_units(self):
"""
Go short the appropriate number of "units" of the
portfolio to open a new position or to close out
a long position.
"""
for i, ticker in enumerate(self.tickers):
if self.weights[i] < 0.0:
self.events_queue.put(
SignalEvent(ticker, "BOT",
int(floor(-1.0 * self.qty * self.weights[i]))))
else:
self.events_queue.put(
SignalEvent(ticker, "SLD",
int(floor(self.qty * self.weights[i]))))
def calculate_signals(self, event):
"""
For a particular received BarEvent, determine whether
it is the end of the month (for that bar) and generate
a liquidation signal, as well as a purchase signal,
for each ticker.
"""
if (event.type in [EventType.BAR, EventType.TICK]
and self._end_of_month(event.time)):
ticker = event.ticker
if self.tickers_invested[ticker]:
liquidate_signal = SignalEvent(ticker, "EXIT")
self.events_queue.put(liquidate_signal)
long_signal = SignalEvent(ticker, "BOT")
self.events_queue.put(long_signal)
self.tickers_invested[ticker] = True
def calculate_signals(self, event):
ticker = self.tickers[0]
if event.type in [EventType.BAR, EventType.TICK
] and event.ticker == ticker:
if not self.invested and self.ticks == 0:
signal = SignalEvent(ticker, "BOT")
self.events_queue.put(signal)
self.invested = True
self.ticks += 1
def calculate_signals(self, event):
"""
Calculate the signals for the strategy.
"""
if event.type == EventType.SENTIMENT:
ticker = event.ticker
if ticker != "SPY":
# Long signal
if (self.invested[ticker] is False
and event.sentiment >= self.sent_buy):
print("LONG %s at %s" % (ticker, event.timestamp))
self.events_queue.put(SignalEvent(ticker, "BOT", self.qty))
self.invested[ticker] = True
# Close signal
if (self.invested[ticker] is True
and event.sentiment <= self.sent_sell):
print("CLOSING LONG %s at %s" % (ticker, event.timestamp))
self.events_queue.put(SignalEvent(ticker, "SLD", self.qty))
self.invested[ticker] = False
def calculate_signals(self, event):
if (event.type in [EventType.BAR, EventType.TICK]
and event.ticker == self.ticker):
if not self.invested and self.bars == 0:
signal = SignalEvent(self.ticker,
"BOT",
suggested_quantity=self.base_quantity)
self.events_queue.put(signal)
self.invested = True
self.bars += 1
def test_create_order_from_signal_basic_check(self):
"""
Tests the "_create_order_from_signal" method
as a basic sanity check.
"""
signal_event = SignalEvent("MSFT", "BOT")
order = self.portfolio_handler._create_order_from_signal(signal_event)
self.assertEqual(order.ticker, "MSFT")
self.assertEqual(order.action, "BOT")
self.assertEqual(order.quantity, 0)
def test_on_signal_basic_check(self):
"""
Tests the "on_signal" method as a basic sanity check.
"""
signal_event = SignalEvent("MSFT", "BOT")
self.portfolio_handler.on_signal(signal_event)
ret_order = self.portfolio_handler.events_queue.get()
self.assertEqual(ret_order.ticker, "MSFT")
self.assertEqual(ret_order.action, "BOT")
self.assertEqual(ret_order.quantity, 100)
def calculate_signals(self, event):
if (event.type == EventType.BAR and event.ticker == self.ticker):
exit_position = False
signal = event.signal[event.signal['symbol'] == event.ticker]
self.signals = self.signals.append(signal)
if self.invested:
# exit_position = self.expired_not_profit()
if not exit_position and len(self.signals) > 2:
two_ago = self.signals.iloc[-2]
one_ago = self.signals.iloc[-1]
exit_position = (
(one_ago['sma'] < one_ago['bb_middleband'])
and (two_ago['sma'] > two_ago['bb_middleband'])
# and len(went_up) < 1
)
if exit_position:
print('SMA exit')
if not exit_position:
entry = self.signals.loc[
self.signals['date'] >= self.entry_time]
# if entry['close'][0] < signal['trailing_stop'][0] and signal['adx'][0] > 10:
if signal['close'][0] < entry['trailing_stop'].max(
) and signal['adx'][0] > 10:
exit_position = True
print('Trailing exit - ' + str(signal['adx'][0]))
# Trading signals based on moving average cross
if signal['buy'][0] == 1 and not self.invested:
print("LONG %s: %s" % (self.ticker, event.time))
self.entry_time = signal['date'][0]
signal_event = SignalEvent(
self.ticker, "BOT", suggested_quantity=self.base_quantity)
self.events_queue.put(signal_event)
self.invested = True
elif (signal['sell'][0] == 1 or exit_position) and self.invested:
print("SHORT %s: %s" % (self.ticker, event.time))
signal_event = SignalEvent(
self.ticker, "SLD", suggested_quantity=self.base_quantity)
self.events_queue.put(signal_event)
self.invested = False
def calculate_signals(self, event):
"""
Calculate the intraday machine learning
prediction strategy.
"""
if event.type == EventType.BAR:
self._update_current_returns(event)
self.minutes += 1
# Allow enough time to pass to populate the
# returns feature vector
if self.minutes > (self.lags + 2):
pred = self.model.predict(self.cur_returns.reshape((1, -1)))[0]
# Long only strategy
if not self.invested and pred == 1:
print("LONG: %s" % event.time)
self.events_queue.put(
SignalEvent(self.tickers[0], "BOT", self.qty))
self.invested = True
if self.invested and pred == -1:
print("CLOSING LONG: %s" % event.time)
self.events_queue.put(
SignalEvent(self.tickers[0], "SLD", self.qty))
self.invested = False
def calculate_signals(self, event):
"""
Calculate the Kalman Filter strategy.
"""
if event.type == EventType.BAR:
self._set_correct_time_and_price(event)
# Only trade if we have both observations
if all(self.latest_prices > -1.0):
# Create the observation matrix of the latest prices
# of TLT and the intercept value (1.0) as well as the
# scalar value of the latest price from IEI
F = np.asarray([self.latest_prices[0], 1.0]).reshape((1, 2))
y = self.latest_prices[1]
# The prior value of the states \theta_t is
# distributed as a multivariate Gaussian with
# mean a_t and variance-covariance R_t
if self.R is not None:
self.R = self.C + self.wt
else:
self.R = np.zeros((2, 2))
# Calculate the Kalman Filter update
# ----------------------------------
# Calculate prediction of new observation
# as well as forecast error of that prediction
yhat = F.dot(self.theta)
et = y - yhat
# Q_t is the variance of the prediction of
# observations and hence \sqrt{Q_t} is the
# standard deviation of the predictions
Qt = F.dot(self.R).dot(F.T) + self.vt
sqrt_Q = np.sqrt(Qt)
# The posterior value of the states \theta_t is
# distributed as a multivariate Gaussian with mean
# m_t and variance-covariance C_t
At = self.R.dot(F.T) / Qt
self.theta = self.theta + At.flatten() * et
self.C = self.R - At * F.dot(self.R)
# Only trade if days is greater than a "burn in" period
if self.days > 1:
# If we’re not in the market...
if self.invested is None:
if e < -sqrt_Q:
# Long Entry
print("LONG: %s" % event.time)
self.cur_hedge_qty = int(
floor(self.qty * self.theta[0]))
self.events_queue.put(
SignalEvent(self.tickers[1], "BOT", self.qty))
self.events_queue.put(
SignalEvent(self.tickers[0], "SLD",
self.cur_hedge_qty))
self.invested = "long"
elif e > sqrt_Q:
# Short Entry
print("SHORT: %s" % event.time)
self.cur_hedge_qty = int(
floor(self.qty * self.theta[0]))
self.events_queue.put(
SignalEvent(self.tickers[1], "SLD", self.qty))
self.events_queue.put(
SignalEvent(self.tickers[0], "BOT", self.qty))
self.invested = "short"
# If we are in the market...
if self.invested is not None:
if self.invested == "long" and e > -sqrt_Q:
print("CLOSING LONG: %s" % event.time)
self.events_queue.put(
SignalEvent(self.tickers[1], "SLD", self.qty))
self.events_queue.put(
SignalEvent(self.tickers[0], "BOT",
self.cur_hedge_qty))
self.invested = None
elif self.invested == "short" and e < sqrt_Q:
print("CLOSING SHORT: %s" % event.time)
self.events_queue.put(
SignalEvent(self.tickers[1], "BOT", self.qty))
self.events_queue.put(
SignalEvent(self.tickers[0], "SLD",
self.cur_hedge_qty))
self.invested = None
if event.type == EventType.BAR:
self._update_current_returns(event)
self.minutes += 1
gap = event.ask - event.bid
# Allow enough time to pass to populate the
# returns feature vector
if self.minutes > (self.lags + 2) and event.close_price > 0: #>0 por seguridad
pred = self.modelpkl.predict(self.cur_returns.reshape((1, -1)))[0]
# Long only strategy
if self.invested == "NONE" and gap <= 200000:
if pred > self.toplimit:
self.qty = int(round(30000/(event.close_price/float(
PriceParser.PRICE_MULTIPLIER)),0)-1.0)
print("LONG: %s" % event.time)
self.events_queue.put(
SignalEvent(self.tickers[0], "buy", self.qty)
)
self.prev_invested = "NONE"
self.invested = "LONG"
self.contador = 0
#self.entry_price = event.close_price/float(
# PriceParser.PRICE_MULTIPLIER)
elif pred < self.botlimit:
self.qty = int(round(30000/(event.close_price/float(
PriceParser.PRICE_MULTIPLIER)),0)-1.0)
print("SHORT: %s" % event.time)
self.events_queue.put(
SignalEvent(self.tickers[0], "sell", self.qty)
)
self.prev_invested = "NONE"
self.invested = "SHORT"
def calculate_signals(self, event):
"""
Calculate the Kalman Filter strategy.
"""
if event.type == EventType.BAR:
self._set_correct_time_and_price(event)
idx = self.tickers.index(event.ticker)
pair_idx = idx // 2
# Only trade if we have both observations
data = self.trader_data[pair_idx]
if not all(data['latest_prices'] > -1.0):
return
F = np.asarray([data['latest_prices'][0], 1.0]).reshape((1, 2))
y = data['latest_prices'][1]
if data['R'] is not None:
data['R'] = data['C'] + data['wt']
else:
data['R'] = np.zeros((2, 2))
yhat = F.dot(data['theta'])
et = y - yhat
Qt = F.dot(data['R']).dot(F.T) + data['vt']
sqrt_Qt = np.sqrt(Qt)
At = data['R'].dot(F.T) / Qt
data['theta'] = data['theta'] + At.flatten() * et
data['C'] = data['R'] - At * F.dot(data['R'])
investment_per_pair = self.investment_per_pair
# if pair_idx == self.num_pairs - 1:
# investment_per_pair *= 0
# continue
if self.days > 1:
# If we're not in the market...
if data['invested'] is None:
if et < -factorA * sqrt_Qt:
# Long Entry
# print("LONG: %s" % event.time)
data['qty'] = floor(investment_per_pair /
data['latest_prices'][1])
data['cur_hedge_qty'] = floor(investment_per_pair /
data['latest_prices'][0])
# data['cur_hedge_qty'] = int(
# floor(data['qty'] * data['theta'][0]))
self.events_queue.put(
SignalEvent(self.tickers[pair_idx * 2 + 1], "BOT",
data['qty']))
self.events_queue.put(
SignalEvent(self.tickers[pair_idx * 2], "SLD",
data['cur_hedge_qty']))
data['invested'] = "long"
elif et > factorA * sqrt_Qt:
# Short Entry
# print("SHORT: %s" % event.time)
data['qty'] = floor(investment_per_pair /
data['latest_prices'][1])
data['cur_hedge_qty'] = floor(investment_per_pair /
data['latest_prices'][0])
# data['cur_hedge_qty'] = int(floor(data['qty'] * data['theta'][0]))
# self.cur_hedge_qty = int(
# floor(self.qty * self.theta[0]))
self.events_queue.put(
SignalEvent(self.tickers[pair_idx * 2 + 1], "SLD",
data['qty']))
self.events_queue.put(
SignalEvent(self.tickers[pair_idx * 2], "BOT",
data['cur_hedge_qty']))
data['invested'] = "short"
# If we are in the market...
if data['invested'] is not None:
if data['invested'] == 'long' and et > -factorC * sqrt_Qt:
# print("CLOSING LONG: %s" % event.time)
self.events_queue.put(
SignalEvent(self.tickers[pair_idx * 2 + 1], "SLD",
data['qty']))
self.events_queue.put(
SignalEvent(self.tickers[pair_idx * 2], "BOT",
data['cur_hedge_qty']))
data['invested'] = None
elif data['invested'] == "short" and et < factorC * sqrt_Qt:
# print("CLOSING SHORT: %s" % event.time)
self.events_queue.put(
SignalEvent(self.tickers[pair_idx * 2 + 1], "BOT",
data['qty']))
self.events_queue.put(
SignalEvent(self.tickers[pair_idx * 2], "SLD",
data['cur_hedge_qty']))
data['invested'] = None
def calculate_signals(self, event):
if self.RF == "RFR":
"""
Calculate the intraday machine learning
prediction strategy.
"""
if event.type == EventType.BAR:
self._update_current_returns(event)
self.minutes += 1
# Allow enough time to pass to populate the
# returns feature vector
#if self.minutes > (self.lags + 2): # original
if self.minutes > (self.lags - 1):
# chk = True
if all(x == self.skip for x in self.delta):
chk = True
else:
chk = False
try:
f = open(
self.current_directory + "/" +
"Discontinuity.txt", "r")
f.close()
f = open(
self.current_directory + "/" +
"Discontinuity.txt", "a")
f.write(str(self.cur_time) + "\n")
f.close()
except IOError:
f = open(
self.current_directory + "/" +
"Discontinuity.txt", "w+")
f.write(str(self.cur_time) + "\n")
f.close()
if chk:
if self.poly >= 2:
cont = 0
for i in range(0, self.lags):
for j in range(i, self.lags):
self.add_returns[cont] = self.cur_returns[
i] * self.cur_returns[j]
cont += 1
if self.poly == 3:
for i in range(0, self.lags):
for j in range(i, self.lags):
for k in range(j, self.lags):
self.add_returns[
cont] = self.cur_returns[
i] * self.cur_returns[
j] * self.cur_returns[k]
cont += 1
cur_returns = np.concatenate(
(self.cur_returns, self.add_returns))
else:
cur_returns = self.cur_returns
try:
f = open(self.current_directory + "/" + "test.txt",
"r")
f.close()
f = open(self.current_directory + "/" + "test.txt",
"a")
f.write(str(event.time) + str(cur_returns) + "\n")
f.close
except IOError:
f = open(self.current_directory + "/" + "test.txt",
"w+")
f.write(str(event.time) + str(cur_returns) + "\n")
f.close
pred = self.modelpkl.predict(
cur_returns.reshape((1, -1)))[0]
#print(event.time, pred)
try:
f = open(
self.current_directory + "/" + "pred_test.txt",
"r")
f.close()
f = open(
self.current_directory + "/" + "pred_test.txt",
"a")
f.write(str(event.time) + "_" + str(pred) + "\n")
f.close
except IOError:
f = open(
self.current_directory + "/" + "pred_test.txt",
"w+")
f.write(str(event.time) + "_" + str(pred) + "\n")
f.close
# current_price = event.close_price/float(
# PriceParser.PRICE_MULTIPLIER)
# Long only strategy
if self.invested == "NONE":
if pred > self.toplimit:
self.qty = int(
round(
30000 /
(event.close_price /
float(PriceParser.PRICE_MULTIPLIER)),
0) - 1.0)
print("LONG: %s" % event.time)
self.events_queue.put(
SignalEvent(self.tickers[0], "BOT",
self.qty))
self.invested = "LONG"
self.contador = 0
#self.entry_price = event.close_price/float(
# PriceParser.PRICE_MULTIPLIER)
# elif pred < self.botlimit:
# self.qty = int(round(30000/(event.close_price/float(
# PriceParser.PRICE_MULTIPLIER)),0)-1.0)
# print("SHORT: %s" % event.time)
# self.events_queue.put(
# SignalEvent(self.tickers[0], "SLD", self.qty)
# )
# self.invested = "SHORT"
# self.contador = 0
# #self.entry_price = event.close_price/float(
# # PriceParser.PRICE_MULTIPLIER)
elif self.invested == "LONG":
self.contador += 1
# if current_price > self.entry_price:
# print("CLOSING LONG: %s" % event.time)
# self.events_queue.put(
# SignalEvent(self.tickers[0], "SLD", self.qty)
# )
# self.invested = "NONE"
#if pred < self.botlimit:
if self.contador == self.salida:
print("CLOSING LONG: %s" % event.time)
self.events_queue.put(
SignalEvent(self.tickers[0], "SLD",
self.qty))
self.invested = "NONE"
# elif self.invested == "SHORT":
# self.contador += 1
# # if current_price < self.entry_price:
# # print("CLOSING LONG: %s" % event.time)
# # self.events_queue.put(
# # SignalEvent(self.tickers[0], "BOT", self.qty)
# # )
# # self.invested = "NONE"
# #if pred > self.toplimit:
# if self.contador == self.salida:
# print("CLOSING SHORT: %s" % event.time)
# self.events_queue.put(
# SignalEvent(self.tickers[0], "BOT", self.qty)
# )
# self.invested = "NONE"
elif self.RF == "RFC":
"""
Calculate the intraday machine learning
prediction strategy.
"""
if event.type == EventType.BAR:
self._update_current_returns(event)
self.minutes += 1
# Allow enough time to pass to populate the
# returns feature vector
if self.minutes > (self.lags + 2):
# chk = True
if all(x == self.skip for x in self.delta):
chk = True
else:
chk = False
try:
f = open(
self.current_directory + "/" +
"Discontinuity.txt", "r")
f.close()
f = open(
self.current_directory + "/" +
"Discontinuity.txt", "a")
f.write(str(self.cur_time) + "\n")
f.close()
except IOError:
f = open(
self.current_directory + "/" +
"Discontinuity.txt", "w+")
f.write(str(self.cur_time) + "\n")
f.close()
if chk:
pred = self.modelpkl.predict(
self.cur_returns.reshape((1, -1)))[0]
if self.invested == "NONE":
if pred == 1:
self.qty = int(
round(
30000 /
(event.close_price /
float(PriceParser.PRICE_MULTIPLIER)),
0) - 1.0)
print("LONG: %s" % event.time)
self.events_queue.put(
SignalEvent(self.tickers[0], "BOT",
self.qty))
self.invested = "LONG"
self.contador = 0
elif pred == -1:
self.qty = int(
round(
30000 /
(event.close_price /
float(PriceParser.PRICE_MULTIPLIER)),
0) - 1.0)
print("SHORT: %s" % event.time)
self.events_queue.put(
SignalEvent(self.tickers[0], "SLD",
self.qty))
self.invested = "SHORT"
self.contador = 0
elif self.invested == "LONG":
self.contador += 1
if self.contador == self.salida:
# if pred == -1:
print("CLOSING LONG: %s" % event.time)
self.events_queue.put(
SignalEvent(self.tickers[0], "SLD",
self.qty))
self.invested = "NONE"
elif self.invested == "SHORT":
self.contador += 1
if self.contador == self.salida:
# if pred == 1:
print("CLOSING SHORT: %s" % event.time)
self.events_queue.put(
SignalEvent(self.tickers[0], "BOT",
self.qty))
self.invested = "NONE"
