class LiveTestStrat():
"""Same as Previous strat but this should use live prices
and trade the same stratey as Strategy()
"""
def __init__(self):
self.HZ = 1.0 / 1.0
self.local_strategies = [self.tsla]
# variables needed for tsla()
self.prices = AutoQueue(10)
self.boundary = 0.01
self.holding = False
self.volume = 0
def execute(self, portfolio):
while True:
for local_strategy in self.local_strategies:
local_strategy()
sleep(1.0 / self.HZ)
def tsla(self):
tsla = Security('TSLA')
price = tsla.get_price()
self.prices.put(price)
volume = tsla.security.get_volume()
# if volume != self.volume:
# print(volume)
# self.volume = volume
# else:
# return
# do things necessary for avg and stddev calc
# self.price_sum = self.prices.sum()
# self.price_sum_sq = self.prices.sum_sq()
# self.count = len(self.prices)
avg = self.prices.average()
var = self.prices.variance()
if price < avg * (1 - self.boundary):
# low price, want to buy
while portfolio.buy(tsla, 1):
pass
self.holding = True
elif price > avg * (1 + self.boundary):
# high price, want to sell_all
portfolio.sell_all(tsla)
self.holding = False
print(
f'Price: {price:.2f}\tAvg: {avg:.2f}\tvar: {var:.2f} <{time.asctime()}',
end='\r')
class Strategy():
def __init__(self):
# TODO framework to test different parameter in parallel
self.HZ = 14000.0
self.price_sum = 0
# perhaps make arg in Q(arg) an arg in __init__
self.prices = AutoQueue(10)
self.price_sum_sq = 0
self.count = 0
self.boundary = 0.025 # percent change from mean
# calculate average using a queue, be able to set size using some parameter
def execute(self, portfolio):
goog = TestSecurity("GOOG")
counter = 0
while counter < 949:
price = goog.get_price()
self.prices.put(price)
# do things necessary for avg and stddev calc
self.price_sum = self.prices.sum()
self.price_sum_sq = self.prices.sum_sq()
self.count = len(self.prices)
avg = self.prices.average()
var = self.prices.variance()
if price < avg * (1 - self.boundary):
# low price, want to buy
while portfolio.buy(goog, 1):
pass
self.holding = True
elif price > avg * (1 + self.boundary):
# high price, want to sell_all
portfolio.sell_all(goog)
self.holding = False
# print(f'Price: {price:.2f}\tAvg: {avg:.2f}\tvar: {var:.2f}')
counter += 1
sleep(1.0 / self.HZ)
print(portfolio)