def add_volatility_indicators(data: pd.DataFrame) -> pd.DataFrame:
"""Adds the volatility indicators.
Parameters
----------
data : pd.DataFrame
A dataframe with daily stock values. Must include: open, high,
low, close and volume. It should also be sorted in a descending
manner.
Returns
-------
pd.DataFrame
The input dataframe with the indicators added.
"""
bb = BollingerBands(data['close'])
data.loc[:, 'bollinger_hband'] = bb.bollinger_hband()
data.loc[:, 'bollinger_hband_indicator'] = bb.bollinger_hband_indicator()
data.loc[:, 'bollinger_lband'] = bb.bollinger_lband()
data.loc[:, 'bollinger_lband_indicator'] = bb.bollinger_lband_indicator()
data.loc[:, 'bollinger_mavg'] = bb.bollinger_mavg()
data.loc[:, 'bollinger_pband'] = bb.bollinger_pband()
data.loc[:, 'bollinger_wband'] = bb.bollinger_wband()
return data
def get_signal(self, input_df, ticker, run_id):
df = input_df.copy()
indicator_bb = BollingerBands(
close=df["Close"],
window=self.indicator["window"],
window_dev=self.indicator["window_dev"],
fillna=self.indicator["fillna"],
)
# Add Bollinger Bands features
df["bb_bbm"] = indicator_bb.bollinger_mavg()
df["bb_bbh"] = indicator_bb.bollinger_hband()
df["bb_bbl"] = indicator_bb.bollinger_lband()
df["bb_bbhi"] = indicator_bb.bollinger_hband_indicator()
df["bb_bbli"] = indicator_bb.bollinger_lband_indicator()
df["bb_bbp"] = indicator_bb.bollinger_pband()
row = df.iloc[-1]
if row.bb_bbhi.item():
sell_signal = {
"ticker":
ticker,
"datetime":
row.Date,
"indicator":
self.name,
"param":
self.param,
"reason":
"High BollingerBand percentage - currently at {}%".format(
int(row.bb_bbp.item() * 100.0)),
"image":
self.draw_image(df, ticker, run_id),
}
else:
sell_signal = None
if row.bb_bbli.item():
buy_signal = {
"ticker":
ticker,
"datetime":
row.Date,
"indicator":
self.name,
"param":
self.param,
"reason":
"Low BollingerBand percentage - currently at {}%".format(
int(row.bb_bbp.item() * 100.0)),
"image":
self.draw_image(df, ticker, run_id),
}
else:
buy_signal = None
return buy_signal, sell_signal
def AddIndicators(df):
# Add Simple Moving Average (SMA) indicators
df["sma7"] = SMAIndicator(close=df["Close"], window=7,
fillna=True).sma_indicator()
df["sma25"] = SMAIndicator(close=df["Close"], window=25,
fillna=True).sma_indicator()
df["sma99"] = SMAIndicator(close=df["Close"], window=99,
fillna=True).sma_indicator()
# Add Bollinger Bands indicator
indicator_bb = BollingerBands(close=df["Close"], window=20, window_dev=2)
df['bb_bbm'] = indicator_bb.bollinger_mavg()
df['bb_bbh'] = indicator_bb.bollinger_hband()
df['bb_bbl'] = indicator_bb.bollinger_lband()
# Add Parabolic Stop and Reverse (Parabolic SAR) indicator
indicator_psar = PSARIndicator(high=df["High"],
low=df["Low"],
close=df["Close"],
step=0.02,
max_step=2,
fillna=True)
df['psar'] = indicator_psar.psar()
# Add Moving Average Convergence Divergence (MACD) indicator
df["MACD"] = macd(close=df["Close"],
window_slow=26,
window_fast=12,
fillna=True) # mazas
# Add Relative Strength Index (RSI) indicator
df["RSI"] = rsi(close=df["Close"], window=14, fillna=True) # mazas
return df
def add_volatility_ta(df: pd.DataFrame,
high: str,
low: str,
close: str,
fillna: bool = False,
colprefix: str = "") -> pd.DataFrame:
"""Add volatility technical analysis features to dataframe.
Args:
df (pandas.core.frame.DataFrame): Dataframe base.
high (str): Name of 'high' column.
low (str): Name of 'low' column.
close (str): Name of 'close' column.
fillna(bool): if True, fill nan values.
colprefix(str): Prefix column names inserted
Returns:
pandas.core.frame.DataFrame: Dataframe with new features.
"""
# Average True Range
df[f'{colprefix}volatility_atr'] = AverageTrueRange(
close=df[close], high=df[high], low=df[low], n=10,
fillna=fillna).average_true_range()
# Bollinger Bands
indicator_bb = BollingerBands(close=df[close], n=20, ndev=2, fillna=fillna)
df[f'{colprefix}volatility_bbm'] = indicator_bb.bollinger_mavg()
df[f'{colprefix}volatility_bbh'] = indicator_bb.bollinger_hband()
df[f'{colprefix}volatility_bbl'] = indicator_bb.bollinger_lband()
df[f'{colprefix}volatility_bbw'] = indicator_bb.bollinger_wband()
df[f'{colprefix}volatility_bbhi'] = indicator_bb.bollinger_hband_indicator(
)
df[f'{colprefix}volatility_bbli'] = indicator_bb.bollinger_lband_indicator(
)
# Keltner Channel
indicator_kc = KeltnerChannel(close=df[close],
high=df[high],
low=df[low],
n=10,
fillna=fillna)
df[f'{colprefix}volatility_kcc'] = indicator_kc.keltner_channel_central()
df[f'{colprefix}volatility_kch'] = indicator_kc.keltner_channel_hband()
df[f'{colprefix}volatility_kcl'] = indicator_kc.keltner_channel_lband()
df[f'{colprefix}volatility_kchi'] = indicator_kc.keltner_channel_hband_indicator(
)
df[f'{colprefix}volatility_kcli'] = indicator_kc.keltner_channel_lband_indicator(
)
# Donchian Channel
indicator_dc = DonchianChannel(close=df[close], n=20, fillna=fillna)
df[f'{colprefix}volatility_dcl'] = indicator_dc.donchian_channel_lband()
df[f'{colprefix}volatility_dch'] = indicator_dc.donchian_channel_hband()
df[f'{colprefix}volatility_dchi'] = indicator_dc.donchian_channel_hband_indicator(
)
df[f'{colprefix}volatility_dcli'] = indicator_dc.donchian_channel_lband_indicator(
)
return df
def AddIndicators(df):
# Add Bollinger Bands indicator
indicator_bb = BollingerBands(close=df["Close"], window=20, window_dev=2)
df['bb_bbm'] = indicator_bb.bollinger_mavg()
df['bb_bbh'] = indicator_bb.bollinger_hband()
df['bb_bbl'] = indicator_bb.bollinger_lband()
return df
class TestBollingerBands(unittest.TestCase):
"""
https://school.stockcharts.com/doku.php?id=technical_indicators:bollinger_bands
"""
_filename = 'ta/tests/data/cs-bbands.csv'
def setUp(self):
self._df = pd.read_csv(self._filename, sep=',')
self._indicator = BollingerBands(close=self._df['Close'],
n=20,
ndev=2,
fillna=False)
def tearDown(self):
del (self._df)
def test_mavg(self):
target = 'MiddleBand'
result = self._indicator.bollinger_mavg()
pd.testing.assert_series_equal(self._df[target].tail(),
result.tail(),
check_names=False)
def test_hband(self):
target = 'HighBand'
result = self._indicator.bollinger_hband()
pd.testing.assert_series_equal(self._df[target].tail(),
result.tail(),
check_names=False)
def test_lband(self):
target = 'LowBand'
result = self._indicator.bollinger_lband()
pd.testing.assert_series_equal(self._df[target].tail(),
result.tail(),
check_names=False)
def test_wband(self):
target = 'WidthBand'
result = self._indicator.bollinger_wband()
pd.testing.assert_series_equal(self._df[target].tail(),
result.tail(),
check_names=False)
def test_hband_indicator(self):
target = 'CrossUp'
result = self._indicator.bollinger_hband_indicator()
pd.testing.assert_series_equal(self._df[target].tail(),
result.tail(),
check_names=False)
def test_lband_indicator(self):
target = 'CrossDown'
result = self._indicator.bollinger_lband_indicator()
pd.testing.assert_series_equal(self._df[target].tail(),
result.tail(),
check_names=False)
def bollinger_bands(df):
indicator_bb = BollingerBands(close=df["Close"])
df['bb_bbh'] = indicator_bb.bollinger_hband()
df['bb_bbl'] = indicator_bb.bollinger_lband()
df['bb_avg'] = indicator_bb.bollinger_mavg()
df['bb_bbh_ind'] = indicator_bb.bollinger_hband_indicator()
df['bb_bbl_ind'] = indicator_bb.bollinger_lband_indicator()
df['bb_pband'] = indicator_bb.bollinger_pband()
df['bb_wband'] = indicator_bb.bollinger_wband()
def bollinger_bands(data, days=20, std=2, price="Close"):
df = data
# Initialize Bollinger Bands Indicator
indicator_bb = BollingerBands(close=df[price], window=days, window_dev=std)
# Add Bollinger Bands features
df["bb_ma"] = indicator_bb.bollinger_mavg()
df["bb_high"] = indicator_bb.bollinger_hband()
df["bb_low"] = indicator_bb.bollinger_lband()
# Add Bollinger Band high indicator
df["bb_bbhi"] = indicator_bb.bollinger_hband_indicator()
# Add Bollinger Band low indicator
df["bb_bbli"] = indicator_bb.bollinger_lband_indicator()
return df
def analysis(df, ma_f, ma_s, period):
df["ema_f"] = ta.ema(high=df.high, low=df.low, close=df.close, length=ma_f)
df["ema_s"] = ta.ema(high=df.high, low=df.low, close=df.close, length=ma_s)
df.fillna(0, inplace=True)
df['s'] = (df['ema_s'] - df['ema_s'].shift(1)) >= 0
df['f'] = (df['ema_f'] - df['ema_f'].shift(1)) >= 0
df['buy_ema'] = df['ema_f'] > df['ema_s']
df['sell_ema'] = df['ema_f'] <= df['ema_s']
df['buy_change'] = (df['buy_ema'] != df['buy_ema'].shift(1)) & df['buy_ema']
df['sell_change'] = (df['sell_ema'] != df['sell_ema'].shift(1)) & df['sell_ema']
# df["RSI"] = ta.rsi(high=df.high, low=df.low, close=df.close, length=period)
# df['RSIs'] = (df['RSI'] - df['RSI'].shift(1)) >= 0
# df['RSI_ups'] = df.groupby(
# (df['RSIs'] != df['RSIs'].shift(1)).cumsum()).cumcount() + 1
# df['adx'] = ta.adx(high=df.high, low=df.low, close=df.close, length=period)['ADX_%s' % period]
# df['adx_s'] = (df['adx'] - df['adx'].shift(1)) >= 0
# df['adx_ups'] = df.groupby(
# (df['adx_s'] != df['adx_s'].shift(1)).cumsum()).cumcount() + 1
# df['ATR'] = df.ta.atr()
indicator_bb = BollingerBands(close=df.close, window=10, window_dev=1.8)
df['bb_bbm'] = indicator_bb.bollinger_mavg()
df['b_m'] = (df['bb_bbm'] - df['bb_bbm'].shift(1)) >= 0
df['bb_bbh'] = indicator_bb.bollinger_hband()
df['bb_bbl'] = indicator_bb.bollinger_lband()
df['bb_bbhi'] = indicator_bb.bollinger_hband_indicator()
df['bb_bbli'] = indicator_bb.bollinger_lband_indicator()
df['pvt'] = ta.pvt(close=df.close, volume=df.volume)
df['pvt_t'] = (df['pvt'] - df['pvt'].shift(1)) >= 0
df['close_variation'] = df['close'] - df['close'].shift(1)
#
# df = ema(df, ma_f, ma_s)
df = rsi(df, period)
df = momentum(df)
df['trend'] = df['momentum_s'] & df['RSIs']
return df
def transform_one(self, entity_id, df: pd.DataFrame) -> pd.DataFrame:
indicator_bb = BollingerBands(close=df["close"],
window=20,
window_dev=2)
# Add Bollinger Bands features
df["bb_bbm"] = indicator_bb.bollinger_mavg()
df["bb_bbh"] = indicator_bb.bollinger_hband()
df["bb_bbl"] = indicator_bb.bollinger_lband()
# Add Bollinger Band high indicator
df["bb_bbhi"] = indicator_bb.bollinger_hband_indicator()
# Add Bollinger Band low indicator
df["bb_bbli"] = indicator_bb.bollinger_lband_indicator()
# Add Width Size Bollinger Bands
df["bb_bbw"] = indicator_bb.bollinger_wband()
# Add Percentage Bollinger Bands
df["bb_bbp"] = indicator_bb.bollinger_pband()
return df
def get_bollinger_bands(config, crypto):
close_prices = crypto.prices
dataframe = crypto.technical_indicators
window = 20
indicator_bb = BollingerBands(close=close_prices,
window=window,
window_dev=2)
dataframe['Bollinger_Bands_Middle'] = indicator_bb.bollinger_mavg()
dataframe['Bollinger_Bands_Upper'] = indicator_bb.bollinger_hband()
dataframe['Bollinger_Bands_Lower'] = indicator_bb.bollinger_lband()
generate_buy_sell_signals(
lambda x, signal: signal['Close'].values[x] < signal[
'Bollinger_Bands_Lower'].values[x], lambda x, signal: signal[
'Close'].values[x] > signal['Bollinger_Bands_Upper'].values[x],
dataframe, 'Bollinger_Bands')
return dataframe
def GetPriceDataFromExchangeFinnHub(event, context):
retVal = {}
retVal["data"] = []
# For debugging the input, write the EVENT to CloudWatch logs
print(json.dumps(event))
# Data is sent to Lambda via a HTTPS POST call. We want to get to the payload send by Snowflake
event_body = event["body"]
payload = json.loads(event_body)
for row in payload["data"]:
sflkRowRef = row[
0] # This is how Snowflake keeps track of data as it gets returned
symbol = row[
1] # The data passed in from Snowflake that the input row contains.
fromDate = row[2]
toDate = row[3]
# Will return URL without token to Snowflake for tracking
URL = f'https://finnhub.io/api/v1/stock/candle?symbol={symbol}&resolution=D&from={fromDate}&to={toDate}'
# Add our FinnHubAPI Key to the end of the URL.
# This is in a new variable which will not be returned to Snowflake
URLWithToken = f'{URL}&token={FinnHubAPI.TOKEN}'
# GET data from the API
httpData = requests.get(url=URLWithToken).json()
# Convert to Pandas DataFrame
df = pd.DataFrame(httpData)
# Add the column names
print("Adding column names")
df.columns = [
"Close", "High", "Low", "Open", "Status", "OpenTime", "Volume"
]
# Set DateTime columns to correct type
df['OpenTime'] = pd.to_datetime(df['OpenTime'], unit='ms')
df['Open'] = df['Open'].astype(float)
df['High'] = df['High'].astype(float)
df['Low'] = df['Low'].astype(float)
df['Close'] = df['Close'].astype(float)
df['Volume'] = df['Volume'].astype(float)
# Clean NaN values
print("Cleaning NA values")
df = dropna(df)
# Calculate the Bollinger Bands indicator
indicator_bb = BollingerBands(close=df["Close"], n=20, ndev=2)
df['bb_bbm'] = indicator_bb.bollinger_mavg()
df['bb_bbh'] = indicator_bb.bollinger_hband()
df['bb_bbl'] = indicator_bb.bollinger_lband()
df['bb_bbhi'] = indicator_bb.bollinger_hband_indicator()
df['bb_bbli'] = indicator_bb.bollinger_lband_indicator()
df['bb_bbw'] = indicator_bb.bollinger_wband()
df['bb_bbp'] = indicator_bb.bollinger_pband()
print("converting OHLC pandas to JSON. This does it as a string")
buffer = df.to_json(orient="records")
print("Interpret the JSON string into a dictionary for output")
jsonResponse = json.loads(buffer)
# Prepare the output response
response = {}
response["url"] = URL
response["response"] = jsonResponse
retVal["data"].append([sflkRowRef, response])
# For debugging the output, write the RETurn VALue to CloudWatch logs
# print(json.dumps(retVal))
return retVal
def handle(self, *args, **options):
# import pdb
# pdb.set_trace()
if not options['update']:
NSETechnical.objects.all().delete()
symbols = Symbol.objects.all()
for symbol in symbols:
nse_history_data = NSEHistoricalData.objects.filter(
symbol__symbol_name=symbol).order_by('timestamp')
if not nse_history_data:
continue
nse_technical = pd.DataFrame(
list(
nse_history_data.values('timestamp', 'open', 'high', 'low',
'close', 'total_traded_quantity')))
'''
Moving average convergence divergence
'''
indicator_macd = MACD(close=nse_technical['close'],
window_slow=26,
window_fast=12,
window_sign=9,
fillna=False)
nse_technical["trend_macd"] = indicator_macd.macd()
nse_technical["trend_macd_signal"] = indicator_macd.macd_signal()
nse_technical["trend_macd_diff"] = indicator_macd.macd_diff()
'''
Simple Moving Average
'''
nse_technical["trend_sma_fast"] = SMAIndicator(
close=nse_technical['close'], window=12,
fillna=False).sma_indicator()
nse_technical["trend_sma_slow"] = SMAIndicator(
close=nse_technical['close'], window=26,
fillna=False).sma_indicator()
'''
Exponential Moving Average
'''
nse_technical["trend_ema_fast"] = EMAIndicator(
close=nse_technical['close'], window=12,
fillna=False).ema_indicator()
nse_technical["trend_ema_slow"] = EMAIndicator(
close=nse_technical['close'], window=26,
fillna=False).ema_indicator()
'''
Ichimoku Indicator
'''
indicator_ichi = IchimokuIndicator(
high=nse_technical['high'],
low=nse_technical['low'],
window1=9,
window2=26,
window3=52,
visual=False,
fillna=False,
)
nse_technical[
"trend_ichimoku_conv"] = indicator_ichi.ichimoku_conversion_line(
)
nse_technical[
"trend_ichimoku_base"] = indicator_ichi.ichimoku_base_line()
nse_technical["trend_ichimoku_a"] = indicator_ichi.ichimoku_a()
nse_technical["trend_ichimoku_b"] = indicator_ichi.ichimoku_b()
indicator_ichi_visual = IchimokuIndicator(
high=nse_technical['high'],
low=nse_technical['low'],
window1=9,
window2=26,
window3=52,
visual=True,
fillna=False,
)
nse_technical[
"trend_visual_ichimoku_a"] = indicator_ichi_visual.ichimoku_a(
)
nse_technical[
"trend_visual_ichimoku_b"] = indicator_ichi_visual.ichimoku_b(
)
'''
Bollinger Band
'''
indicator_bb = BollingerBands(close=nse_technical['close'],
window=20,
window_dev=2,
fillna=False)
nse_technical["volatility_bbm"] = indicator_bb.bollinger_mavg()
nse_technical["volatility_bbh"] = indicator_bb.bollinger_hband()
nse_technical["volatility_bbl"] = indicator_bb.bollinger_lband()
nse_technical["volatility_bbw"] = indicator_bb.bollinger_wband()
nse_technical["volatility_bbp"] = indicator_bb.bollinger_pband()
nse_technical[
"volatility_bbhi"] = indicator_bb.bollinger_hband_indicator()
nse_technical[
"volatility_bbli"] = indicator_bb.bollinger_lband_indicator()
'''
Accumulation Distribution Index
'''
nse_technical["volume_adi"] = AccDistIndexIndicator(
high=nse_technical['high'],
low=nse_technical['low'],
close=nse_technical['close'],
volume=nse_technical['total_traded_quantity'],
fillna=False).acc_dist_index()
'''
Money Flow Index
'''
nse_technical["volume_mfi"] = MFIIndicator(
high=nse_technical['high'],
low=nse_technical['low'],
close=nse_technical['close'],
volume=nse_technical['total_traded_quantity'],
window=14,
fillna=False,
).money_flow_index()
'''
Relative Strength Index (RSI)
'''
nse_technical["momentum_rsi"] = RSIIndicator(
close=nse_technical['close'], window=14, fillna=False).rsi()
'''
Stoch RSI (StochRSI)
'''
indicator_srsi = StochRSIIndicator(close=nse_technical['close'],
window=14,
smooth1=3,
smooth2=3,
fillna=False)
nse_technical["momentum_stoch_rsi"] = indicator_srsi.stochrsi()
nse_technical["momentum_stoch_rsi_k"] = indicator_srsi.stochrsi_k()
nse_technical["momentum_stoch_rsi_d"] = indicator_srsi.stochrsi_d()
nse_technical.replace({np.nan: None}, inplace=True)
nse_technical.replace([np.inf, -np.inf], None, inplace=True)
list_to_create = []
list_to_update = []
for index in range(len(nse_history_data) - 1, -1, -1):
data = nse_history_data[index]
if data.technicals:
break
technical = NSETechnical(
nse_historical_data=data,
trend_macd=nse_technical['trend_macd'][index],
trend_macd_signal=nse_technical['trend_macd_signal']
[index],
trend_macd_diff=nse_technical['trend_macd_diff'][index],
trend_sma_fast=nse_technical['trend_sma_fast'][index],
trend_sma_slow=nse_technical['trend_sma_slow'][index],
trend_ema_fast=nse_technical['trend_ema_fast'][index],
trend_ema_slow=nse_technical['trend_ema_slow'][index],
trend_ichimoku_conv=nse_technical['trend_ichimoku_conv']
[index],
trend_ichimoku_base=nse_technical['trend_ichimoku_base']
[index],
trend_ichimoku_a=nse_technical['trend_ichimoku_a'][index],
trend_ichimoku_b=nse_technical['trend_ichimoku_b'][index],
trend_visual_ichimoku_a=nse_technical[
'trend_visual_ichimoku_a'][index],
trend_visual_ichimoku_b=nse_technical[
'trend_visual_ichimoku_b'][index],
volatility_bbm=nse_technical['volatility_bbm'][index],
volatility_bbh=nse_technical['volatility_bbh'][index],
volatility_bbl=nse_technical['volatility_bbl'][index],
volatility_bbw=nse_technical['volatility_bbw'][index],
volatility_bbp=nse_technical['volatility_bbp'][index],
volatility_bbhi=nse_technical['volatility_bbhi'][index],
volatility_bbli=nse_technical['volatility_bbli'][index],
volume_adi=nse_technical['volume_adi'][index],
volume_mfi=nse_technical['volume_mfi'][index],
momentum_rsi=nse_technical['momentum_rsi'][index],
momentum_stoch_rsi=nse_technical['momentum_stoch_rsi']
[index],
momentum_stoch_rsi_k=nse_technical['momentum_stoch_rsi_k']
[index],
momentum_stoch_rsi_d=nse_technical['momentum_stoch_rsi_d']
[index])
data.technicals = True
list_to_update.append(data)
list_to_create.append(technical)
NSETechnical.objects.bulk_create(list_to_create)
NSEHistoricalData.objects.bulk_update(list_to_update,
['technicals'])
print(f"Technicals updated for {symbol}")
def _bollinger_bands(self, df, close):
bb = BollingerBands(close=close)
df['bb_bbm'] = bb.bollinger_mavg()
df['bb_bbh'] = bb.bollinger_hband()
df['bb_bbl'] = bb.bollinger_lband()
features_considered = [
'Close', 'Volume', 'MA_short', 'MA_long', 'Wilders_EMA', 'bb_bbm',
'bb_bbh', 'bb_bbl', 'bb_bbhi', 'bb_bbli'
]
# features_considered = ['Close', 'MA_short', 'MA_long', 'Wilders_EMA']
# Initialize Bollinger Bands Indicator
e = EquityData('data/SPY.csv', 'SPY')
indicator_bb = BollingerBands(close=e.close(), n=20, ndev=2)
e.data['MA_short'] = moving_average(e, window=history_size)
e.data['MA_long'] = moving_average(e, window=5)
e.data['Wilders_EMA'] = e.close().ewm(alpha=1 / history_size,
adjust=False).mean()
# Add Bollinger Bands features
e.data['bb_bbm'] = indicator_bb.bollinger_mavg()
e.data['bb_bbh'] = indicator_bb.bollinger_hband()
e.data['bb_bbl'] = indicator_bb.bollinger_lband()
# Add Bollinger Band high indicator
e.data['bb_bbhi'] = indicator_bb.bollinger_hband_indicator()
# Add Bollinger Band low indicator
e.data['bb_bbli'] = indicator_bb.bollinger_lband_indicator()
e.data = e.data[21:]
EVALUATION_INTERVAL = int(e.data.shape[0] / BATCH_SIZE) * 2
features = e.data[features_considered]
assert (list(features)[0] == 'Close')
features.index = e.date()
# features.plot(subplots=True)
# plt.show()
def add_volatility_ta(
df: pd.DataFrame,
high: str,
low: str,
close: str,
fillna: bool = False,
colprefix: str = "",
vectorized: bool = False,
) -> pd.DataFrame:
"""Add volatility technical analysis features to dataframe.
Args:
df (pandas.core.frame.DataFrame): Dataframe base.
high (str): Name of 'high' column.
low (str): Name of 'low' column.
close (str): Name of 'close' column.
fillna(bool): if True, fill nan values.
colprefix(str): Prefix column names inserted
vectorized(bool): if True, use only vectorized functions indicators
Returns:
pandas.core.frame.DataFrame: Dataframe with new features.
"""
# Bollinger Bands
indicator_bb = BollingerBands(close=df[close],
window=20,
window_dev=2,
fillna=fillna)
df[f"{colprefix}volatility_bbm"] = indicator_bb.bollinger_mavg()
df[f"{colprefix}volatility_bbh"] = indicator_bb.bollinger_hband()
df[f"{colprefix}volatility_bbl"] = indicator_bb.bollinger_lband()
df[f"{colprefix}volatility_bbw"] = indicator_bb.bollinger_wband()
df[f"{colprefix}volatility_bbp"] = indicator_bb.bollinger_pband()
df[f"{colprefix}volatility_bbhi"] = indicator_bb.bollinger_hband_indicator(
)
df[f"{colprefix}volatility_bbli"] = indicator_bb.bollinger_lband_indicator(
)
# Keltner Channel
indicator_kc = KeltnerChannel(close=df[close],
high=df[high],
low=df[low],
window=10,
fillna=fillna)
df[f"{colprefix}volatility_kcc"] = indicator_kc.keltner_channel_mband()
df[f"{colprefix}volatility_kch"] = indicator_kc.keltner_channel_hband()
df[f"{colprefix}volatility_kcl"] = indicator_kc.keltner_channel_lband()
df[f"{colprefix}volatility_kcw"] = indicator_kc.keltner_channel_wband()
df[f"{colprefix}volatility_kcp"] = indicator_kc.keltner_channel_pband()
df[f"{colprefix}volatility_kchi"] = indicator_kc.keltner_channel_hband_indicator(
)
df[f"{colprefix}volatility_kcli"] = indicator_kc.keltner_channel_lband_indicator(
)
# Donchian Channel
indicator_dc = DonchianChannel(high=df[high],
low=df[low],
close=df[close],
window=20,
offset=0,
fillna=fillna)
df[f"{colprefix}volatility_dcl"] = indicator_dc.donchian_channel_lband()
df[f"{colprefix}volatility_dch"] = indicator_dc.donchian_channel_hband()
df[f"{colprefix}volatility_dcm"] = indicator_dc.donchian_channel_mband()
df[f"{colprefix}volatility_dcw"] = indicator_dc.donchian_channel_wband()
df[f"{colprefix}volatility_dcp"] = indicator_dc.donchian_channel_pband()
if not vectorized:
# Average True Range
df[f"{colprefix}volatility_atr"] = AverageTrueRange(
close=df[close],
high=df[high],
low=df[low],
window=10,
fillna=fillna).average_true_range()
# Ulcer Index
df[f"{colprefix}volatility_ui"] = UlcerIndex(
close=df[close], window=14, fillna=fillna).ulcer_index()
return df
from keras.layers import Conv1D, MaxPool1D, Bidirectional, LSTM, Dropout, TimeDistributed
from keras.layers import Dense, GlobalAveragePooling2D
from ta.trend import IchimokuIndicator
from sklearn.linear_model import LinearRegression
from ta import add_all_ta_features
from ta.utils import dropna
import matplotlib.pyplot as plt
filename = 'AAPL'
stock = pd.read_csv('Data/' + filename + '.csv')
indicator_bb = BollingerBands(close=stock["Close"], n=20, ndev=2)
macd = MACD(close=stock["Close"])
rsi = RSIIndicator(close=stock["Close"])
ichi = IchimokuIndicator(high=stock["High"], low=stock["Low"])
stock['macd'] = macd.macd()
stock['rsi'] = rsi.rsi()
stock['bb_bbm'] = indicator_bb.bollinger_mavg()
stock['bb_bbh'] = indicator_bb.bollinger_hband()
stock['bb_bbl'] = indicator_bb.bollinger_lband()
stock['ichi_a'] = ichi.ichimoku_a()
stock['ichi_b'] = ichi.ichimoku_b()
stock['ichi_base'] = ichi.ichimoku_base_line()
stock['ichi_conv'] = ichi.ichimoku_conversion_line()
stock = stock.fillna(0)
print(stock)
scaler = preprocessing.MinMaxScaler()
scaled_values = scaler.fit_transform(stock.iloc[:, 1:4])
stock.iloc[:, 1:4] = scaled_values
y_scaler = preprocessing.MinMaxScaler()
scaled_values = y_scaler.fit_transform(
np.array(stock.iloc[:, 4]).reshape(-1, 1))
df['ppo'] = ppo_indicator.ppo()
roc_indicator = ROCIndicator(close = df["Close"], window = 12, fillna = False)
df['roc'] = roc_indicator.roc()
macd_indicator = MACD(close = df["Close"], window_slow = 20, window_fast = 12, window_sign = 9, fillna = False)
df['macd'] = macd_indicator.macd()
rsi_indicator = RSIIndicator(close = df["Close"], window = 14, fillna = False)
df['rsi'] = rsi_indicator.rsi()
aroon_indicator = AroonIndicator(close = df["Close"], window = 20, fillna = False)
df['aroon'] = aroon_indicator.aroon_indicator()
boll_indicator = BollingerBands(close = df["Close"], window = 20, window_dev = 2, fillna = False)
df['boll_mavg'] = boll_indicator.bollinger_mavg()
df.rename(columns = {"Close": "price"}, inplace=True)
prices = df['price'].to_numpy()
df['day_of_month'] = df.index.day
df['day_of_week'] = df.index.dayofweek
df['month'] = df.index.month
df.dropna(inplace=True)
df = df.drop(df.columns[[0, 1, 2, 4, 5]], axis=1)
X_columns = ['price', 'kama', 'ppo', 'roc', 'macd', 'rsi', 'aroon', 'boll_mavg',
'day_of_month', 'day_of_week', 'month']
X_data = df.filter(X_columns)
figure, axis = plt.subplots(4, 3)
row = 0
column = 0
# Loop over the tickers
for ticker in stock_data_biotech.columns:
# Initalise the DataFrame
data_plot = pd.DataFrame(stock_data_biotech[ticker])
# Initialize Bollinger Bands Indicator
indicator_bb = BollingerBands(close=stock_data_biotech[ticker],
window=20,
window_dev=2)
# Add Bollinger Bands features
data_plot['bb_bbm'] = indicator_bb.bollinger_mavg()
data_plot['bb_bbh'] = indicator_bb.bollinger_hband()
data_plot['bb_bbl'] = indicator_bb.bollinger_lband()
# Create the plot
axis[row, column].plot(data_plot)
axis[row, column].set_title(health_etfs_in_biotech[ticker]['long_name'],
fontsize=6)
axis[row, column].set_xticks([])
axis[row, column].set_yticks([])
# Count a column further
column += 1
# If the column reaches the max, add another row
if column == 3: