def get_features(self, prices):
"""
Compute the technical features of a position and feed that
into the Q-Learning process
TODO: Try different parameters to see if you can improve
the portfolio
Try out different window sizes (right now set to 10 trading days)
prices: Adjusted close prices of the given symbol
df_features: A pandas dataframe of the technical indicators
"""
rolling_mean = prices.rolling(window=self.window_size).mean()
rolling_std = prices.rolling(window=self.window_size).std()
momentum = get_momentum(prices, self.window_size)
sma_indicator = get_sma_indicator(prices, rolling_mean)
bollinger_val = compute_bollinger_value(prices, rolling_mean, rolling_std)
df_features = pd.concat([momentum, sma_indicator], axis=1)
df_features = pd.concat([df_features, bollinger_val], axis=1)
df_features.columns = ["ind{}".format(i)
for i in range(len(df_features.columns))]
# Perhaps you can come up with a different feature?
# Try it!
df_features.dropna(inplace=True)
return df_features
def get_features(self, prices):
"""Compute technical indicators and use them as features to be fed
into a Q-learner.
Parameters:
prices: Adjusted close prices of the given symbol
Returns:
df_features: A pandas dataframe of the technical indicators
"""
window = 10
# Compute rolling mean
rolling_mean = prices.rolling(window=window).mean()
# Compute_rolling_std
rolling_std = prices.rolling(window=window).std()
# Compute momentum
momentum = get_momentum(prices, window)
# Compute SMA indicator
sma_indicator = get_sma_indicator(prices, rolling_mean)
# Compute Bollinger value
bollinger_val = compute_bollinger_value(prices, rolling_mean, rolling_std)
# Create a dataframe with three features
df_features = pd.concat([momentum, sma_indicator], axis=1)
df_features = pd.concat([df_features, bollinger_val], axis=1)
df_features.columns = ["ind{}".format(i)
for i in range(len(df_features.columns))]
df_features.dropna(inplace=True)
return df_features
def trading_strategy(self, sym_price):
"""Create a dataframe of order signals that maximizes portfolio's return.
This function has been optimized for the symbol and training period in
the main function
Parameters:
sym_price: The price series of a stock symbol of interest
Returns:
df_order_signals: A series that contains 1 for buy, 0 for hold and -1 for sell
"""
# Get momentum indicator and generate signals
momentum = get_momentum(sym_price, 40)
mom_signal = -1 * (momentum < -0.07) + 1 * (momentum > 0.14)
# Get RSI indicator and generate signals
rsi_indicator = get_RSI(sym_price)
rsi_signal = 1 * (rsi_indicator < 50) + -1 * (rsi_indicator > 50)
# Get SMA indicator and generate signals
sma_indicator = get_sma_indicator(sym_price,
sym_price.rolling(window=30).mean())
sma_signal = 1 * (sma_indicator < 0.0) + -1 * (sma_indicator > 0.0)
# Combine individual signals
signal = 1 * ((sma_signal == 1) & (rsi_signal == 1) & (mom_signal == 1)) \
+ -1 * ((sma_signal == -1) & (rsi_signal == -1) & (mom_signal == -1))
# Create an order series with 0 as default values
self.df_order_signals = signal * 0
# Keep track of net signals which are constrained to -1, 0, and 1
net_signals = 0
for date in self.df_order_signals.index:
net_signals = self.df_order_signals.loc[:date].sum()
# If net_signals is not long and signal is to buy
if (net_signals < 1) and (signal.loc[date] == 1):
self.df_order_signals.loc[date] = 1
# If net_signals is not short and signal is to sell
elif (net_signals > -1) and (signal.loc[date] == -1):
self.df_order_signals.loc[date] = -1
# On the last day, close any open positions
if self.df_order_signals.sum() == -1:
self.df_order_signals[-1] = 1
elif self.df_order_signals.sum() == 1:
self.df_order_signals[-1] = -1
return self.df_order_signals
def get_features1(self, prices, symbol, print=False):
'''
Compute technical indicators and use them as features to be fed
into a Q-learner.
:param: prices: Adj Close prices dataframe
:param: Whether adding data for printing to dataframe or not
:return: Normalize Features dataframe
'''
# Fill NAN values if any
prices.fillna(method="ffill", inplace=True)
prices.fillna(method="bfill", inplace=True)
prices.fillna(1.0, inplace=True)
# Price
adj_close = prices[symbol]
# Compute Momentum
mom = get_momentum(adj_close, window=10)
# Compute RSI
rsi = get_RSI(adj_close)
# Compute rolling mean
rm = get_rolling_mean(adj_close, window=10)
# Compute rolling standard deviation
rstd = get_rolling_std(adj_close, window=10)
# Compute upper and lower bands
upper_band, lower_band = get_bollinger_bands(rm, rstd)
# Compute SMA
sma = get_sma_indicator(adj_close, rm)
df = prices.copy()
df['Momentum'] = mom
df['Adj. Close/SMA'] = adj_close/sma
df['middle_band'] = rm
# Delete 'Adj Close' column
del df[symbol]
if set(['cash']).issubset(df.columns):
del df['cash']
# Normalize dataframe
df_norm = normalize_data(df)
# Add Adj Close, Bollinrt Bands and RSY if printing
if print:
df_norm['Adj Close'] = prices[symbol]
df_norm['upper_band'] = upper_band
df_norm['lower_band'] = lower_band
df_norm['middle_band'] = rm
df_norm['RSI'] = rsi
# Drop NaN
df_norm.dropna(inplace=True)
return df_norm