def _compute_indicators(prices, highs, lows, volumes, lookback):
mtm = momentum(prices, lookback)
sma, sma_ratio = simple_moving_average(prices, lookback)
bbands = bollinger_bands(prices, sma, lookback)
mfi = money_flow_index(prices, highs, lows, volumes, lookback)
return mtm, sma_ratio, bbands, mfi
def addEvidence(self, symbol = "IBM", \
sd=dt.datetime(2008,1,1), \
ed=dt.datetime(2009,1,1), \
sv = 10000):
window_size = self.window_size
num_feature = self.idt_size
num_n_day_return = 10
threshold = max(0.03, 2 * self.impact)
prices = ut.get_data([symbol], pd.date_range(sd, ed))
prices = prices[symbol]
sma = indi.simple_moving_average(prices, window_size=window_size)
X = []
Y = []
for i in range(window_size + num_feature + 1,
len(prices) - num_n_day_return):
X.append(np.array(sma[i - num_feature:i]))
gain = (prices.values[i + num_n_day_return] -
prices.values[i]) / prices.values[i]
if gain > threshold:
Y.append(1)
elif gain < -threshold:
Y.append(-1)
else:
Y.append(0)
X = np.array(X)
Y = np.array(Y)
self.learner.addEvidence(X, Y)
def testPolicy(symbol='JPM',
sd=dt.datetime(2008, 1, 1),
ed=dt.datetime(2009, 12, 31),
sv=100000):
prices = get_data([symbol], pd.date_range(sd, ed))
prices = prices[symbol]
dates = pd.date_range(sd, ed)
prices_all = get_data([symbol], dates) # automatically adds SPY
df_trades = prices_all[[
symbol,
]].copy(deep=True) # only portfolio symbols
# df_trades = pd.DataFrame(data=np.zeros(len(prices.index)), index=prices.index, columns = ['val'])
df_trades.values[:, :] = 0
current = 0
my_window = 20
BB = bollinger_band(prices, window=my_window, show_pic=False)
SMA = simple_moving_average(prices, window=my_window, show_pic=False)
EMA = exponential_moving_average(prices, window=my_window, show_pic=False)
# print [x for x in BB if x > 1 or x < -1]
for i in range(my_window, len(prices.index)):
if SMA[i] < -0.1:
df_trades.values[i, :] = 1000 - current
current = 1000
elif SMA[i] > 0.1:
df_trades.values[i, :] = -1000 - current
current = -1000
return df_trades
def main():
'''main function'''
#reads in list of stock dates, their corresponding prices and stock symbol
dates, prices, symbol = _make_valid_request()
#reads in signal strategy and the number of days back to assess
signal_strategy = _get_signal_strategy()
days = _get_number_of_days_back()
#finds list of averages and list of strategies for simple moving average
if signal_strategy == 'Simple Moving Average':
data = indicators.simple_moving_average(days, prices).execute()
strategies = signal_strategies.moving_average_signal(prices,
data).execute()
#prints header and report data for simple moving average
report = _create_report_header(symbol, signal_strategy, days)
report += format_report(signal_strategy, dates, prices, data,
strategies)
#finds list of directional indicators and strategies for directional indicators
elif signal_strategy == 'Directional':
positive_threshold = _get_valid_threshold('positive')
negative_threshold = _get_valid_threshold('negative')
data = indicators.directional_indicator(days, prices).execute()
strategies = signal_strategies.directional_signal(
prices, data, positive_threshold, negative_threshold).execute()
#prints header and report data for directional indicators
report = _create_report_header(symbol, signal_strategy, days,
positive_threshold, negative_threshold)
report += format_report(signal_strategy, dates, prices, data,
strategies)
print(report)
def testPolicy(symbol='JPM',
sd=dt.datetime(2008, 1, 1),
ed=dt.datetime(2009, 12, 31),
sv=100000):
prices = get_data([symbol], pd.date_range(sd, ed))
prices = prices[symbol]
df_trades = pd.DataFrame(data=np.zeros(len(prices.index)),
index=prices.index,
columns=['val'])
current = 0
my_window = 20
BB = bollinger_band(prices, window=my_window, show_pic=False)
SMA = simple_moving_average(prices, window=my_window, show_pic=False)
EMA = exponential_moving_average(prices, window=my_window, show_pic=False)
# Manual Strategy using Simple Moving Average
for i in range(my_window, len(prices.index)):
# Smaller than threshold -> buy
# Bigger than threshold -> sell
if SMA[i] < -0.1:
df_trades['val'].iloc[i] = 1000 - current
current = 1000
elif SMA[i] > 0.1:
df_trades['val'].iloc[i] = -current - 1000
current = -1000
return df_trades
def main():
'''main function'''
#reads in list of stock dates, their corresponding prices and stock symbol
dates,prices,symbol = _make_valid_request()
#reads in signal strategy and the number of days back to assess
signal_strategy = _get_signal_strategy()
days = _get_number_of_days_back()
#finds list of averages and list of strategies for simple moving average
if signal_strategy == 'Simple Moving Average':
data = indicators.simple_moving_average(days,prices).execute()
strategies = signal_strategies.moving_average_signal(prices, data).execute()
#prints header and report data for simple moving average
report = _create_report_header(symbol, signal_strategy, days)
report += format_report(signal_strategy,dates,prices,data,strategies)
#finds list of directional indicators and strategies for directional indicators
elif signal_strategy == 'Directional':
positive_threshold = _get_valid_threshold('positive')
negative_threshold = _get_valid_threshold('negative')
data = indicators.directional_indicator(days,prices).execute()
strategies = signal_strategies.directional_signal(prices, data,positive_threshold,negative_threshold).execute()
#prints header and report data for directional indicators
report = _create_report_header(symbol,signal_strategy, days, positive_threshold,negative_threshold)
report += format_report(signal_strategy,dates,prices,data,strategies)
print(report)
def main():
'''main function'''
while True:
symbol = input('Please enter the ticker symbol of the stock you want to search: ')
print()
while True:
start_date = input('Please enter the start date of the analysis in the following format:\nYYYY-MM-DD\n')
if _D_.check_date(start_date) == True:
if _D_.date_to_days(start_date) <= _D_.date_of_today():
break
print("Error. The start date should be on or before today's date. Please try again.")
print()
print()
while True:
end_date = input('Please enter the end date:\nYYYY-MM-DD\n')
if _D_.check_date(end_date) == True:
if _D_.date_to_days(start_date) <= _D_.date_to_days(end_date):
if _D_.date_to_days(end_date) <= _D_.date_of_today():
break
else:
print("Error. The end date should be on or before today's date. Please try again.")
else:
print('Error. The end date should be later than the start date. Please try again.')
print()
if webwork.data(symbol, start_date, end_date) != False:
data_list = webwork.data(symbol, start_date, end_date)
command = input(Menu)
if command.upper() == 'A':
N = int(input('Please enter the range of moving average: '))
indicator = indicators.simple_moving_average(data_list, N)
signal = signals.sm_signal(data_list, N, indicator.cal())
print()
print('SYMBOL: {}\nSTRATEGY: Simple moving average ({}-day)\n'.format(symbol, N))
print_information_s(data_list, N, indicator.cal(), signal.cal())
break
elif command.upper() == 'B':
N = int(input('Please enter the range of directional indicator: '))
while True:
buy_threshold = int(input('Please enter the buy threshold: '))
sell_threshold = int(input('Please enter the sell threshold: '))
if buy_threshold > sell_threshold:
break
else:
print('Error. The buy threshold should be larger than the sell threshold. Please try again.\n')
indicator = indicators.directional(data_list, N)
signal = signals.dir_signal(indicator.cal(), buy_threshold, sell_threshold)
print()
print('SYMBOL: {}\nSTRATEGY: Directional ({}-day), buy above {}, sell below {}\n'.format(symbol, N, '%+d'%buy_threshold, sell_threshold))
print_information_d(data_list, N, indicator.cal(), signal.cal())
break
else:
invalid_command(command)
print('Please try again.\n')
def _compute_indicators(self):
price = self._stock_data.price
high = self._stock_data.high
low = self._stock_data.low
volume = self._stock_data.volume
mtm = momentum(price, self._lookback)
sma, sma_ratio = simple_moving_average(price, self._lookback)
bbands = bollinger_bands(price, sma, self._lookback)
mfi = money_flow_index(price, high, low, volume, self._lookback)
self._indicators = self._discretize((mtm, sma_ratio, bbands, mfi))
def testPolicy(self, symbol = "IBM", \
sd=datetime.datetime(2009,1,1), \
ed=datetime.datetime(2010,1,1), \
sv = 10000):
current_holding = 0
dates = pd.date_range(sd, ed)
prices_all = ut.get_data([symbol], dates) # automatically adds SPY
trades = prices_all[[
symbol,
]].copy(deep=True) # only portfolio symbols
trades_SPY = prices_all['SPY'] # only SPY, for comparison later
window_size = self.window_size
feature_size = self.feature_size
prices = ut.get_data([symbol], pd.date_range(sd, ed))
prices = prices[symbol]
SMA = simple_moving_average(prices, window=window_size, show_pic=False)
BB = bollinger_band(prices, window=window_size, show_pic=False)
EMA = exponential_moving_average(prices,
window=window_size,
show_pic=False)
trades.values[:, :] = 0
Xtest = []
for i in range(window_size + feature_size + 1, len(prices) - 1):
data = np.concatenate(
(SMA[i - feature_size:i], BB[i - feature_size:i],
EMA[i - feature_size:i]))
Xtest.append(data)
res = self.learner.query(Xtest)
for i, r in enumerate(res):
if r > 0:
# Buy signal
trades.values[i + window_size + feature_size +
1, :] = 1000 - current_holding
current_holding = 1000
elif r < 0:
# Sell signal
trades.values[i + window_size + feature_size +
1, :] = -1000 - current_holding
current_holding = -1000
if self.verbose: print type(trades) # it better be a DataFrame!
if self.verbose: print trades
if self.verbose: print prices_all
return trades
def testPolicy(self,
symbol="IBM",
sd=dt.datetime(2009, 1, 1),
ed=dt.datetime(2010, 1, 1),
sv=10000):
current_holding = 0
dates = pd.date_range(sd, ed)
prices_all = ut.get_data([symbol], dates)
trades = prices_all[[
symbol,
]].copy(deep=True)
window_size = self.window_size
num_feature = self.idt_size
prices = ut.get_data([symbol], pd.date_range(sd, ed))
prices = prices[symbol]
# Indicator 1: simple moving average
indi_val = indi.simple_moving_average(prices, window_size=window_size)
# Indicator 2: bollinger_band
#indi_val = indi.bollinger_band(prices, window_size=window_size)
# Indicator 3: momentum
#indi_val = indi.momentum(prices, window_size=window_size)
trades.values[:, :] = 0
Xtest = []
for i in range(window_size + num_feature + 1, len(prices) - 1):
data = np.array(indi_val[i - num_feature:i])
Xtest.append(data)
result = self.learner.query(Xtest)
for i, r in enumerate(result):
if r > 0:
trades.values[i + window_size + num_feature +
1, :] = 1000 - current_holding
current_holding = 1000
elif r < 0:
trades.values[i + window_size + num_feature +
1, :] = -1000 - current_holding
current_holding = -1000
return trades
def execute(self, stocks:list) -> dict:
''' Generates a BUY or SELL signal and returns the result in
a dictionary.
'''
sma_indicator = indicators.simple_moving_average(self._days)
stock_indicators = sma_indicator.execute(stocks)
signals = {}
for index in range(len(stocks)):
if index < self._days:
signals[stocks[index].Date] = ''
elif stock_indicators[stocks[index - 1].Date] != '':
if (float(stocks[index].Close) > float(stock_indicators[stocks[index].Date])) and (float(stocks[index - 1].Close) < float(stock_indicators[stocks[index - 1].Date])):
signals[stocks[index].Date] = 'BUY'
elif (float(stocks[index].Close) < float(stock_indicators[stocks[index].Date])) and (float(stocks[index - 1].Close) > float(stock_indicators[stocks[index - 1].Date])):
signals[stocks[index].Date] = 'SELL'
else:
signals[stocks[index].Date] = ''
else:
signals[stocks[index].Date] = ''
return signals
def testPolicy(symbol = 'JPM', sd = dt.datetime(2008, 1, 1), ed = dt.datetime(2009, 12, 31), sv = 100000):
prices = get_data([symbol], pd.date_range(sd, ed))
prices = prices[symbol]
df_trades = pd.DataFrame(data=np.zeros(len(prices.index)), index=prices.index, columns = ['val'])
current = 0
window_size = 20
sma = indi.simple_moving_average(prices, window_size=window_size)
for i in range(window_size, len(prices.index)):
# the threshold here is 0.1
# Smaller than threshold -> buy
# Bigger than threshold -> sell
if sma[i] < -0.1:
df_trades['val'].iloc[i] = 1000 - current
current = 1000
elif sma[i] > 0.1:
df_trades['val'].iloc[i] = - current - 1000
current = -1000
return df_trades
def addEvidence(self, symbol = "IBM", \
sd=datetime.datetime(2008,1,1), \
ed=datetime.datetime(2009,1,1), \
sv = 10000):
window_size = self.window_size
feature_size = self.feature_size
# N day return (predict the N-day return)
N = self.N
# threshold to determine whether it's a buy signal or a sell signal or nothing
threshold = max(0.03, 2 * self.impact)
prices = ut.get_data([symbol], pd.date_range(sd, ed))
prices = prices[symbol]
SMA = simple_moving_average(prices, window=window_size, show_pic=False)
BB = bollinger_band(prices, window=window_size, show_pic=False)
EMA = exponential_moving_average(prices,
window=window_size,
show_pic=False)
X = []
Y = []
for i in range(window_size + feature_size + 1, len(prices) - N):
# X will be a feature_size * 3 dimension data (concatenate three indicators as feature)
X.append(
np.concatenate(
(SMA[i - feature_size:i], BB[i - feature_size:i],
EMA[i - feature_size:i])))
ret = (prices.values[i + N] - prices.values[i]) / prices.values[i]
if ret > threshold:
Y.append(1)
elif ret < -threshold:
Y.append(-1)
else:
Y.append(0)
X = np.array(X)
Y = np.array(Y)
self.learner.addEvidence(X, Y)
def menu() -> None:
'''Menu for program
'''
while True:
symbol = _get_symbol()
start_date = _get_start_date()
end_date = _get_end_date(start_date)
strategy = _strategy_choice()
days = _get_days()
buy = 0
sell = 0
try:
stocks = download_quotes.get_quotes(symbol, start_date, end_date)
break
except:
print('Could not download stock quotes!')
indicator = ''
signal = ''
indicator_result = ''
signal_result = ''
if strategy == 'Directional':
print('Buy threshold')
buy = _get_threshold()
print('Sell threshold')
sell = _get_threshold()
indicator = indicators.directional(days)
signal = signals.directional(days, buy, sell)
indicator_result = indicator.execute(stocks)
signal_result = signal.execute(stocks)
else:
indicator = indicators.simple_moving_average(days)
signal = signals.simple_moving_average(days)
indicator_result = indicator.execute(stocks)
signal_result = signal.execute(stocks)
_print_report(symbol, strategy, days, buy, sell, stocks, indicator_result,
signal_result)
return None
def menu() -> None:
'''Menu for program
'''
while True:
symbol = _get_symbol()
start_date = _get_start_date()
end_date = _get_end_date(start_date)
strategy = _strategy_choice()
days =_get_days()
buy = 0
sell = 0
try:
stocks = download_quotes.get_quotes(symbol, start_date, end_date)
break
except:
print('Could not download stock quotes!')
indicator = ''
signal = ''
indicator_result = ''
signal_result = ''
if strategy == 'Directional':
print('Buy threshold')
buy = _get_threshold()
print('Sell threshold')
sell = _get_threshold()
indicator = indicators.directional(days)
signal = signals.directional(days, buy, sell)
indicator_result = indicator.execute(stocks)
signal_result = signal.execute(stocks)
else:
indicator = indicators.simple_moving_average(days)
signal = signals.simple_moving_average(days)
indicator_result = indicator.execute(stocks)
signal_result = signal.execute(stocks)
_print_report(symbol, strategy, days, buy, sell, stocks, indicator_result, signal_result)
return None