def save_all_indicators(stock_name):
'''
Saves everything to local .csv files with corresponding name
@DEV NOTE: can only process any 5 of them, the other 2 will encounter unexpected error
:param stock_name:
:return: None
'''
ti = TechIndicators(key=Env.alpha_vantage_api_key, output_format='pandas')
d, _ = ti.get_cci(symbol=stock_name, interval='daily', time_period=60)
d.to_csv(stock_name + '_CCI.csv', index=True, sep=',')
d, _ = ti.get_rsi(symbol=stock_name, interval='daily', series_type='close', time_period='60')
d.to_csv(stock_name + '_RSI.csv', index=True, sep=',')
d, _ = ti.get_stoch(symbol=stock_name, interval='daily')
d.to_csv(stock_name + '_stoch.csv', index=True, sep=',')
d, _ = ti.get_macd(symbol=stock_name, interval='daily', series_type='close')
d.to_csv(stock_name + '_MACD.csv', index=True, sep=',')
d, _ = ti.get_obv(symbol=stock_name, interval='daily')
d.to_csv(stock_name + '_OBV.csv', index=True, sep=',')
d, _ = ti.get_ema(symbol=stock_name, interval='daily', series_type='close')
d.to_csv(stock_name + '_EMA.csv', index=True, sep=',')
d, _ = ti.get_roc(symbol=stock_name, interval='daily', time_period=60, series_type='close')
d.to_csv(stock_name + '_ROC.csv', index=True, sep=',')
def get_data(symbol):
# Technical Indicators
ti = TechIndicators(key='YOUR_API_KEY', output_format='pandas')
sma, _ = ti.get_sma(symbol=symbol, interval='daily')
wma, _ = ti.get_wma(symbol=symbol, interval='daily')
ema, _ = ti.get_ema(symbol=symbol, interval='daily')
macd, _ = ti.get_macd(symbol=symbol, interval='daily')
stoch, _ = ti.get_stoch(symbol=symbol, interval='daily')
rsi, _ = ti.get_rsi(symbol=symbol, interval='daily')
adx, _ = ti.get_adx(symbol=symbol, interval='daily')
cci, _ = ti.get_cci(symbol=symbol, interval='daily')
aroon, _ = ti.get_aroon(symbol=symbol, interval='daily')
bbands, _ = ti.get_bbands(symbol=symbol, interval='daily')
ad, _ = ti.get_ad(symbol=symbol, interval='daily')
obv, _ = ti.get_obv(symbol=symbol, interval='daily')
mom, _ = ti.get_mom(symbol=symbol, interval='daily')
willr, _ = ti.get_willr(symbol=symbol, interval='daily')
tech_ind = pd.concat([sma, ema, macd, stoch, rsi, adx, cci, aroon, bbands, ad, obv, wma, mom, willr], axis=1)
ts = TimeSeries(key='YOUR_API_KEY', output_format='pandas')
close = ts.get_daily(symbol=symbol, outputsize='full')[0]['close'] # compact/full
direction = (close > close.shift()).astype(int)
target = direction.shift(-1).fillna(0).astype(int)
target.name = 'target'
data = pd.concat([tech_ind, close, target], axis=1)
return data
def get_daily_technical(stock, indicator, API_key, period=-1):
ti = TechIndicators(key=API_key, output_format='pandas')
if indicator == "bband":
if (period <= 0):
period = 20
data, meta_data = ti.get_bbands(symbol=stock, interval='daily', time_period=period)
elif indicator == "macd":
data, meta_data = ti.get_macd(symbol=stock, interval='daily')
elif indicator == "rsi":
if (period <= 0):
period = 14
data, meta_data = ti.get_rsi(symbol=stock, interval='daily', time_period=period)
elif indicator == "cci":
if (period <= 0):
period = 20
data, meta_data = ti.get_cci(symbol=stock, interval='daily', time_period=period)
elif indicator == "aroon":
if (period <= 0):
period = 14
data, meta_data = ti.get_aroon(symbol=stock, interval='daily', time_period=period)
elif indicator == "ad":
data, meta_data = ti.get_ad(symbol=stock, interval='daily')
elif indicator == "adx":
if (period <= 0):
period = 20
data, meta_data = ti.get_adx(symbol=stock, interval='daily', time_period=period)
elif indicator == "sma":
if (period <= 0):
period = 40
data, meta_data = ti.get_sma(symbol=stock, interval='daily', time_period=period)
else:
sys.exit('Failed to input a valid indicator')
return data, meta_data
def get_data(symbol):
print 'Getting data'
# Technical Indicators
ti = TechIndicators(key='4BTFICZGTPWZRRQS', output_format='pandas')
sma, _ = ti.get_sma(symbol=symbol, interval='daily')
wma, _ = ti.get_wma(symbol='SPX', interval='daily')
ema, _ = ti.get_ema(symbol=symbol, interval='daily')
macd, _ = ti.get_macd(symbol=symbol, interval='daily')
stoch, _ = ti.get_stoch(symbol=symbol, interval='daily')
rsi, _ = ti.get_rsi(symbol=symbol, interval='daily')
adx, _ = ti.get_adx(symbol=symbol, interval='daily')
cci, _ = ti.get_cci(symbol=symbol, interval='daily')
aroon, _ = ti.get_aroon(symbol=symbol, interval='daily')
bbands, _ = ti.get_bbands(symbol='SPX', interval='daily')
ad, _ = ti.get_ad(symbol='SPX', interval='daily')
obv, _ = ti.get_obv(symbol='SPX', interval='daily')
mom, _ = ti.get_mom(symbol='SPX', interval='daily')
willr, _ = ti.get_willr(symbol='SPX', interval='daily')
tech_ind = pd.concat([sma, ema, macd, stoch, rsi, adx, cci, aroon, bbands, ad, obv, wma, mom, willr], axis=1)
print 'Getting time series'
ts = TimeSeries(key='4BTFICZGTPWZRRQS', output_format='pandas')
close = ts.get_daily(symbol=symbol, outputsize='full')[0].rename(columns={'4. close': 'close'})['close']
direction = (close > close.shift()).astype(int)
target = direction.shift(-1).fillna(0).astype(int)
target.name = 'target'
data = pd.concat([tech_ind, close, target], axis=1)
return data
def get_alpha(ticker_list, alpha_key):
'''
Input:
ticker_list: list of strings of publicly traded companies symbols, 'XYZ'
alpha_key: get alpha vantage API key from alpha vantage website
Output:
build CSV files of stock price history, and technical indicators for
each company
'''
ts = TimeSeries(key=alpha_key, output_format='pandas')
ti = TechIndicators(key=alpha_key, output_format='pandas')
zero_time = time()
# Build CSV for Daily prices, RSI, CCI, OBV for each company
for ticker in ticker_list:
print('company: {}, time: {}'.format(ticker, str(time() - zero_time)))
df, meta = ts.get_daily(symbol=ticker, outputsize='full')
df.to_csv('data/daily/{}.csv'.format(ticker))
sleep(1)
df, meta = ti.get_rsi(ticker, interval='daily', time_period=5)
df.to_csv('data/rsi/{}.csv'.format(ticker))
sleep(1)
df, meta = ti.get_cci(ticker, interval='daily', time_period=5)
df.to_csv('data/cci/{}.csv'.format(ticker))
sleep(1)
df, meta = ti.get_obv(ticker, interval='daily')
df.to_csv('data/obv/{}.csv'.format(ticker))
def CCI(stock_name):
'''
Commodity Channel Index (顺势指标)
https://en.wikipedia.org/wiki/Commodity_channel_index
https://baike.baidu.com/item/%E9%A1%BA%E5%8A%BF%E6%8C%87%E6%A0%87/976711
'''
ti = TechIndicators(key=Env.alpha_vantage_api_key, output_format='pandas')
data, meta_data = ti.get_cci(symbol=stock_name, interval='daily', time_period=60)
# data.to_csv(stock_name + '_CCI indicator.csv', index=True, sep=',')
return data.to_json(orient='split')
def get_data(symbol):
# API KEY: 51N1JMO66W56AYA3
ti = TechIndicators(key='51N1JMO66W56AYA3', output_format='pandas')
sma, _ = ti.get_sma(symbol=symbol, interval='daily')
wma, _ = ti.get_wma(symbol=symbol, interval='daily')
ema, _ = ti.get_ema(symbol=symbol, interval='daily')
macd, _ = ti.get_macd(symbol=symbol, interval='daily')
stoch, _ = ti.get_stoch(symbol=symbol, interval='daily')
# Alpha Vantage Times out for more than 5 request
time.sleep(65)
rsi, _ = ti.get_rsi(symbol=symbol, interval='daily')
adx, _ = ti.get_adx(symbol=symbol, interval='daily')
cci, _ = ti.get_cci(symbol=symbol, interval='daily')
aroon, _ = ti.get_aroon(symbol=symbol, interval='daily')
bbands, _ = ti.get_bbands(symbol=symbol, interval='daily')
time.sleep(65)
ad, _ = ti.get_ad(symbol=symbol, interval='daily')
obv, _ = ti.get_obv(symbol=symbol, interval='daily')
mom, _ = ti.get_mom(symbol=symbol, interval='daily')
willr, _ = ti.get_willr(symbol=symbol, interval='daily')
time.sleep(65)
tech_ind = pd.concat([
sma, ema, macd, rsi, adx, cci, aroon, bbands, ad, obv, wma, mom, willr,
stoch
],
axis=1)
ts = TimeSeries(key='51N1JMO66W56AYA3', output_format='pandas')
close2 = ts.get_daily(symbol=symbol, outputsize='full')
close = ts.get_daily(symbol=symbol, outputsize='full')[0]['4. close']
direction = (close > close.shift()).astype(int)
target = direction.shift(-1).fillna(0).astype(int)
target.name = 'target'
data = pd.concat([tech_ind, close, target], axis=1)
return data
def data_to_csv(ticker, dire):
# Your key here - keep it - not an issue, free
key = '6LGG0QGAGBROB2M6'
# make the directory if it doesn't exist
if os.path.isdir(dire) == False:
os.mkdir(ticker)
os.mkdir(dire)
ts = TimeSeries(key=key, output_format='pandas')
ti = TechIndicators(key=key, output_format='pandas')
### PRICE ###############################################################
intra, meta = ts.get_intraday(symbol=ticker,
interval='1min',
outputsize='full')
intra.to_csv(dire + ticker + "_prices.csv")
### SMA #################################################################
sma8, meta = ti.get_sma(symbol=ticker, interval='1min', time_period=8)
sma8.to_csv(dire + ticker + "_sma8.csv")
### EMA #################################################################
ema8, meta = ti.get_ema(symbol=ticker, interval='1min', time_period=8)
ema8.to_csv(dire + ticker + "_ema8.csv")
### VWAP ################################################################
vwap, meta = ti.get_vwap(symbol=ticker, interval='1min')
vwap.to_csv(dire + ticker + "_vwap.csv")
### MACD ################################################################
macd, meta = ti.get_macd(symbol=ticker, interval='1min')
macd.to_csv(dire + ticker + "_macd.csv")
# they limit user to 5 calls per minute and 500 per day for free access #
print("waiting so we don't overload API (60 seconds...)")
time.sleep(60)
### STOCH ###############################################################
stoch, meta = ti.get_stoch(symbol=ticker, interval='1min')
stoch.to_csv(dire + ticker + "_stoch.csv")
### RSI #################################################################
rsi, meta = ti.get_rsi(symbol=ticker, interval='1min', time_period=60)
rsi.to_csv(dire + ticker + "_rsi.csv")
### ADX #################################################################
adx, meta = ti.get_adx(symbol=ticker, interval='1min', time_period=60)
adx.to_csv(dire + ticker + "_adx.csv")
### CCI #################################################################
cci, meta = ti.get_cci(symbol=ticker, interval='1min', time_period=60)
cci.to_csv(dire + ticker + "_cci.csv")
# TODO: These are tricky, several lines, parse differently, separate indices really
### AROON ###############################################################
aroon, meta = ti.get_aroon(symbol=ticker, interval='1min', time_period=60)
aroon.to_csv(dire + ticker + "_aroon.csv")
# they limit user to 5 calls per minute and 500 per day for free access #
print("waiting so we don't overload API (60 seconds...)")
time.sleep(60)
### BBANDS ###############################################################
bbands, meta = ti.get_bbands(symbol=ticker,
interval='1min',
time_period=60)
bbands.to_csv(dire + ticker + "_bbands.csv")
'''
def stockIndicators(self):
ti = TechIndicators(key=self.apiKey, output_format='pandas')
if (self.stockIndi == "sma"):
ts = TimeSeries(key=self.apiKey, output_format='pandas')
data, meta_data = ts.get_monthly(symbol=self.stockName)
dataIndi, meta_dataIndi = ti.get_sma(symbol=self.stockName,
time_period=30)
elif (self.stockIndi == "ema"):
ts = TimeSeries(key=self.apiKey, output_format='pandas')
data, meta_data = ts.get_monthly(symbol=self.stockName)
dataIndi, meta_dataIndi = ti.get_ema(symbol=self.stockName,
time_period=30)
elif (self.stockIndi == "vwap"):
ts = TimeSeries(key=self.apiKey, output_format='pandas')
data, meta_data = ts.get_monthly(symbol=self.stockName)
dataIndi, meta_dataIndi = ti.get_vwap(symbol=self.stockName,
time_period=30)
elif (self.stockIndi == "macd"):
ts = TimeSeries(key=self.apiKey, output_format='pandas')
data, meta_data = ts.get_monthly(symbol=self.stockName)
dataIndi, meta_dataIndi = ti.get_macd(symbol=self.stockName,
time_period=30)
elif (self.stockIndi == "obv"):
ts = TimeSeries(key=self.apiKey, output_format='pandas')
data, meta_data = ts.get_monthly(symbol=self.stockName)
dataIndi, meta_dataIndi = ti.get_obv(symbol=self.stockName,
time_period=30)
elif (self.stockIndi == "ad"):
ts = TimeSeries(key=self.apiKey, output_format='pandas')
data, meta_data = ts.get_monthly(symbol=self.stockName)
dataIndi, meta_dataIndi = ti.get_ad(symbol=self.stockName,
time_period=30)
elif (self.stockIndi == "bbands"):
ts = TimeSeries(key=self.apiKey, output_format='pandas')
data, meta_data = ts.get_monthly(symbol=self.stockName)
dataIndi, meta_dataIndi = ti.get_bbands(symbol=self.stockName,
time_period=30)
elif (self.stockIndi == "aroon"):
ts = TimeSeries(key=self.apiKey, output_format='pandas')
data, meta_data = ts.get_monthly(symbol=self.stockName)
dataIndi, meta_dataIndi = ti.get_aroon(symbol=self.stockName,
time_period=30)
elif (self.stockIndi == "cci"):
ts = TimeSeries(key=self.apiKey, output_format='pandas')
data, meta_data = ts.get_monthly(symbol=self.stockName)
dataIndi, meta_dataIndi = ti.get_cci(symbol=self.stockName,
time_period=30)
elif (self.stockIndi == "adx"):
ts = TimeSeries(key=self.apiKey, output_format='pandas')
data, meta_data = ts.get_monthly(symbol=self.stockName)
dataIndi, meta_dataIndi = ti.get_adx(symbol=self.stockName,
time_period=30)
elif (self.stockIndi == "rsi"):
ts = TimeSeries(key=self.apiKey, output_format='pandas')
data, meta_data = ts.get_monthly(symbol=self.stockName)
dataIndi, meta_dataIndi = ti.get_rsi(symbol=self.stockName,
time_period=30)
elif (self.stockIndi == "stoch"):
ts = TimeSeries(key=self.apiKey, output_format='pandas')
data, meta_data = ts.get_monthly(symbol=self.stockName)
dataIndi, meta_dataIndi = ti.get_stoch(symbol=self.stockName,
time_period=30)
return data, dataIndi
def get_daily_history(token_1, symbol, amount_of_entries=1000):
ts_1 = TimeSeries(key=token_1, output_format='pandas')
ti_1 = TechIndicators(key=token_1, output_format='pandas')
print('get_daily_history')
interval = 'daily'
series_type = 'open'
# Fetch dataset from api
# OHLC series #1
time_series, meta_time_series = ts_1.get_daily(symbol=symbol, outputsize='full')
# Bollinger Bands #2
bbands, meta = ti_1.get_bbands(symbol=symbol, interval=interval, time_period=60, series_type=series_type)
bbands = bbands.sort_index(ascending=False)
# SMA10 or MA10 #3
sma_10, meta = ti_1.get_sma(symbol=symbol, interval=interval, time_period=10, series_type=series_type)
sma_10 = sma_10.sort_index(ascending=False)
# SMA5 or MA5 #4
sma_5, meta = ti_1.get_sma(symbol=symbol, interval=interval, time_period=5, series_type=series_type)
sma_5 = sma_5.sort_index(ascending=False)
# ROC #5
roc_20, meta = ti_1.get_roc(symbol=symbol, interval=interval, time_period=20, series_type=series_type)
roc_20 = roc_20.sort_index(ascending=False)
# Wait 70 seconds
print('waiting 70 seconds');
sleep(70)
# MACD #1
macd, meta = ti_1.get_macd(symbol=symbol, interval=interval, series_type=series_type)
macd = macd.sort_index(ascending=False)
# CCI #2
cci_20, meta = ti_1.get_cci(symbol=symbol, interval=interval, time_period=20)
cci_20 = cci_20.sort_index(ascending=False)
# ATR #3
atr_20, meta = ti_1.get_atr(symbol=symbol, interval=interval, time_period=20)
atr_20 = atr_20.sort_index(ascending=False)
# EMA20 #4
ema_20, meta = ti_1.get_ema(symbol=symbol, interval=interval, time_period=20, series_type=series_type)
ema_20 = ema_20.sort_index(ascending=False)
# MTM6 #5
mtm_6, meta = ti_1.get_mom(symbol=symbol, interval=interval, time_period=180, series_type=series_type)
mtm_6 = mtm_6.sort_index(ascending=False)
# Wait 70 seconds
print('waiting 70 seconds');
sleep(70)
# MTM12 #1
mtm_12, meta = ti_1.get_mom(symbol=symbol, interval=interval, time_period=360, series_type=series_type)
mtm_12 = mtm_12.sort_index(ascending=False)
# Get last n data points in the dataset and set your respective column name
d1 = time_series[:amount_of_entries];
d1.columns = ['open', 'high', 'low', 'close', 'volume']
d2 = sma_10[:amount_of_entries]
d2.columns = ['sma10']
d3 = sma_5[:amount_of_entries]
d3.columns = ['sma5']
d4 = pd.DataFrame(bbands['Real Middle Band'][:amount_of_entries])
d4.columns = ['bbands']
d5 = roc_20[:amount_of_entries]
d5.columns = ['roc']
d6 = pd.DataFrame(macd['MACD'][:amount_of_entries])
d6.columns = ['macd']
d7 = cci_20[:amount_of_entries]
d7.columns = ['cci']
d8 = atr_20[:amount_of_entries]
d8.columns = ['atr']
d9 = ema_20[:amount_of_entries]
d9.columns = ['ema20']
d10 = mtm_6[:amount_of_entries]
d10.columns = ['mtm6']
d11 = mtm_12[:amount_of_entries]
d11.columns = ['mtm12']
# Merge elements
merged = d1.merge(d2, left_index=True, right_index=True)
merged = merged.merge(d3, left_index=True, right_index=True)
merged = merged.merge(d4, left_index=True, right_index=True)
merged = merged.merge(d5, left_index=True, right_index=True)
merged = merged.merge(d6, left_index=True, right_index=True)
merged = merged.merge(d7, left_index=True, right_index=True)
merged = merged.merge(d8, left_index=True, right_index=True)
merged = merged.merge(d9, left_index=True, right_index=True)
merged = merged.merge(d10, left_index=True, right_index=True)
merged = merged.merge(d11, left_index=True, right_index=True)
sep = os.path.sep
filename = 'datasets'+sep+'history_'+symbol+'.csv'
merged.to_csv(filename, index=True)
return merged, meta_time_series
sma, meta_data_sma = ti.get_sma(symbol=stock, interval=t_period, series_type= series_type) print sma.shape #5 rsi, meta_data_rsi = ti.get_rsi(symbol=stock, interval=t_period, series_type= series_type) print rsi.shape #6 macd, meta_data_macd = ti.get_macd(symbol=stock,interval=t_period, series_type= series_type) print macd.shape #7 stoch, meta_data_stoch = ti.get_stoch(symbol=stock, interval=t_period) print stoch.shape #8 adx, meta_data_adx = ti.get_adx(symbol=stock, interval=t_period) print adx.shape #9 cci, meta_data_cci = ti.get_cci(symbol=stock, interval=t_period) print cci.shape #10 aroon, meta_data_aroon = ti.get_aroon(symbol=stock, interval=t_period, series_type=series_type) print aroon.shape #11 ad, meta_data_ad = ti.get_ad(symbol=stock, interval=t_period) print ad.shape #12 obv, meta_data_obv = ti.get_obv(symbol=stock, interval=t_period) print obv.shape #13 mom, meta_data_mom = ti.get_mom(symbol=stock, interval=t_period, series_type= series_type) #14 willr, meta_data_willr = ti.get_willr(symbol=stock, interval=t_period)
def get_technical(symbol, is_save=False):
"""
Esta función es para descargar todos los indicadores técnicos.
Si usamos una API_KEY gratuita de ALPHA VANTAGE tenemos que
descomentar los time.sleep() de la función para que no de un
problema de peticiones; en el caso de una API_KEY premium no haría
falta descomentarlo.
Parámetros:
----------
symbol str:
Nombre de la acción en bolsa.
is_save bool:
Booleano para decidir si guardar o no los datos
descargados. Por defecto False no guarda los
datos descargados
"""
try:
# Comprueba si ya existe o no el archivo y en el caso de que
# si exista guarda solo los días que no estén descargados
if os.path.isfile(f"{c.PATH_SAVE_TECH}{symbol}.csv"):
df = pd.read_csv(f"{c.PATH_SAVE_TECH}{symbol}.csv",
names=c.COLUMNS)
df = df[df['symbol'] == symbol]
df['date'] = pd.to_datetime(df['date'])
ld = df['date'].tail(1)
last = datetime(ld.dt.year, ld.dt.month, ld.dt.day)
else:
last = None
techindc = list()
# Descarga los datos de indicadores técnicos.
ti = TechIndicators(key=c.ALPHA_VAN_KEY, output_format='pandas')
init = time.time()
macd = ti.get_macd(symbol, interval='daily')[0]
techindc.append(macd)
stoch = ti.get_stoch(symbol, interval='daily')[0]
techindc.append(stoch)
ad = ti.get_ad(symbol, interval='daily')[0]
techindc.append(ad)
obv = ti.get_obv(symbol, interval='daily')[0]
techindc.append(obv)
# time.sleep(60)
sma = ti.get_sma(symbol, interval='daily', time_period=50)[0]
sma.columns = [f"{c}50" for c in sma.columns]
techindc.append(sma)
bbands = ti.get_bbands(symbol, interval='daily', time_period=28)[0]
bbands.columns = [f"{c}28" for c in bbands.columns]
techindc.append(bbands)
for tp in c.TIME_PERIOD:
rsi = ti.get_rsi(symbol, interval='daily', time_period=tp)[0]
rsi.columns = [f"{c}{tp}" for c in rsi.columns]
techindc.append(rsi)
adx = ti.get_adx(symbol, interval='daily', time_period=tp)[0]
adx.columns = [f"{c}{tp}" for c in adx.columns]
techindc.append(adx)
# time.sleep(60)
cci = ti.get_cci(symbol, interval='daily', time_period=tp)[0]
cci.columns = [f"{c}{tp}" for c in cci.columns]
techindc.append(cci)
aroon = ti.get_aroon(symbol, interval='daily', time_period=tp)[0]
aroon.columns = [f"{c}{tp}" for c in aroon.columns]
techindc.append(aroon)
df_techindc = pd.concat(techindc, axis=1, join='inner')
df_techindc.reset_index(inplace=True)
df_techindc['symbol'] = symbol
if last is not None:
df_techindc = df_techindc[df_techindc['date'] > last]
# Guardar los datos
if is_save:
df_techindc[c.COLUMNS_TECH].to_csv(
f"{c.PATH_SAVE_TECH}{symbol}.csv",
mode='a',
index=False,
header=False)
except:
LOGGER.warning(f"Ticker {symbol} ha fallado.")
def init_data(companies):
# pd.DataFrame(),pd.DataFrame(), pd.DataFrame(),[],pd.DataFrame(),
for comp in companies:
# main_data[comp] = pd.read_csv(f'./rsc/{comp}_data.csv', sep = ',')
# main_data[comp].dropna()
# Input all the technical indicators and the stock times series data
ti = TechIndicators(key='L2O2TYTG382ETN0N', output_format='pandas')
# Fetch the simple moving average (SMA) values
main_data[comp], meta_data = ti.get_sma(symbol=comp,
interval='1min',
time_period=100,
series_type='close')
# Fetch the exponential moving average (EMA) values
ema_data, meta_data = ti.get_ema(symbol=comp,
interval='1min',
time_period=100,
series_type='close')
main_data[comp]['EMA'] = ema_data
"""
# Fetch the weighted moving average (WMA) values
wma_data, meta_data = ti.get_wma(symbol=comp,interval='1min', time_period=10, series_type='close')
main_data[comp]['WMA'] = wma_data
# Fetch the double exponential moving agerage (DEMA) values
dema_data, meta_data = ti.get_dema(symbol=comp,interval='1min', time_period=10, series_type='close')
main_data[comp]['DEMA'] = dema_data
# Fetch the triple exponential moving average (TEMA) values
tema_data, meta_data = ti.get_tema(symbol=comp,interval='1min', time_period=10, series_type='close')
main_data[comp]['TEMA'] = tema_data
# Fetch the triangular moving average (TRIMA) values
trima_data, meta_data = ti.get_trima(symbol=comp,interval='1min', time_period=10, series_type='close')
main_data[comp]['TRIMA'] = trima_data
# Fetch the Kaufman adaptive moving average (KAMA) values
kama_data, meta_data = ti.get_kama(symbol=comp,interval='1min', time_period=10, series_type='close')
main_data[comp]['KAMA'] = kama_data
# Fetch the MESA adaptive moving average (MAMA) values
mama_data, meta_data = ti.get_mama(symbol=comp,interval='1min', time_period=10, series_type='close')
main_data[comp]['MAMA'] = mama_data['MAMA']
main_data[comp]['FAMA'] = mama_data['FAMA']
# Fetch the triple exponential moving average (T3) values
t3_data, meta_data = ti.get_t3(symbol=comp,interval='1min', time_period=10, series_type='close')
main_data[comp]['T3'] = t3_data
"""
# Fetch the moving average convergence / divergence (MACD) values
macd_data, meta_data = ti.get_macd(symbol=comp,
interval='1min',
series_type='close')
main_data[comp]['MACD'] = macd_data['MACD']
main_data[comp]['MACD_Hist'] = macd_data['MACD_Hist']
main_data[comp]['MACD_Signal'] = macd_data['MACD_Signal']
"""
# Fetch the moving average convergence / divergence values with controllable moving average type
macdext_data, meta_data = ti.get_macdext(symbol=comp,interval='1min', series_type='close')
main_data[comp]['MACDEXT'] = macdext_data['MACD']
main_data[comp]['MACDEXT_Hist'] = macdext_data['MACD_Hist']
main_data[comp]['MACDEXT_Signal'] = macdext_data['MACD_Signal']
"""
# Fetch the stochastic oscillator (STOCH) values
stoch_data, meta_data = ti.get_stoch(symbol=comp, interval='1min')
main_data[comp]['SlowK'] = stoch_data['SlowK']
main_data[comp]['SlowD'] = stoch_data['SlowD']
"""
# Fetch the stochastic fast (STOCHF) values
stochf_data, meta_data = ti.get_stochf(symbol=comp,interval='1min')
main_data[comp]['FastK'] = stochf_data['FastK']
main_data[comp]['FastD'] = stochf_data['FastD']
"""
# Fetch the relative strength index (RSI) values
rsi_data, meta_data = ti.get_rsi(symbol=comp,
interval='1min',
time_period=10,
series_type='close')
main_data[comp]['RSI'] = rsi_data
"""
# Fetch the stochastic relative strength index (STOCHRSI) values
stochrsi_data, meta_data = ti.get_stochrsi(symbol=comp,interval='1min', time_period=10, series_type='close')
main_data[comp]['STOCHRSI_FastK'] = stochrsi_data['FastK']
main_data[comp]['STOCHRSI_FastD'] = stochrsi_data['FastD']
# Fetch the Williams' %R (WILLR) values
willr_data, meta_data = ti.get_willr(symbol=comp,interval='1min', time_period=10)
main_data[comp]['WILLR'] = willr_data
"""
# Fetch the average directional movement index (ADX) values
adx_data, meta_data = ti.get_adx(symbol=comp,
interval='1min',
time_period=100)
main_data[comp]['ADX'] = adx_data
"""
# Fetch the average directional movement index rating (ADXR) values
adxr_data, meta_data = ti.get_adxr(symbol=comp,interval='1min', time_period=10)
main_data[comp]['ADXR'] = adxr_data
# Fetch the absolute price oscillator (APO) values
apo_data, meta_data = ti.get_apo(symbol=comp,interval='1min', series_type='close')
main_data[comp]['APO'] = apo_data
# Fetch the percentage price oscillator (PPO) values
ppo_data, meta_data = ti.get_ppo(symbol=comp,interval='1min', series_type='close')
main_data[comp]['PPO'] = ppo_data
# Fetch the momentum (MOM) values
mom_data, meta_data = ti.get_mom(symbol=comp,interval='1min', time_period=10, series_type='close')
main_data[comp]['MOM'] = mom_data
# Fetch the balance of power (BOP) values
bop_data, meta_data = ti.get_bop(symbol=comp,interval='1min')
main_data[comp]['BOP'] = bop_data
"""
time.sleep(5)
# Fetch the commodity channel index (CCI) values
cci_data, meta_data = ti.get_cci(symbol=comp,
interval='1min',
time_period=100)
main_data[comp]['CCI'] = cci_data
"""
# Fetch the Chande momentum oscillator (CMO) values
cmo_data, meta_data = ti.get_cmo(symbol=comp,interval='1min', time_period=10, series_type='close')
main_data[comp]['CMO'] = cmo_data
# Fetch the rate of change (ROC) values
roc_data, meta_data = ti.get_roc(symbol=comp,interval='1min', time_period=10, series_type='close')
main_data[comp]['ROC'] = roc_data
# Fetch the rate of change ratio (ROCR) values
rocr_data, meta_data = ti.get_rocr(symbol=comp,interval='1min', time_period=10, series_type='close')
main_data[comp]['ROCR'] = rocr_data
time.sleep(5)
"""
# Fetch the Aroon (AROON) values
aroon_data, meta_data = ti.get_aroon(symbol=comp,
interval='1min',
time_period=100)
main_data[comp]['Aroon Down'] = aroon_data['Aroon Down']
main_data[comp]['Aroon Up'] = aroon_data['Aroon Up']
"""
# Fetch the Aroon oscillator (AROONOSC) values
aroonosc_data, meta_data = ti.get_aroonosc(symbol=comp,interval='1min', time_period=10)
main_data[comp]['AROONOSC'] = aroonosc_data
# Fetch the money flow index (MFI) values
mfi_data, meta_data = ti.get_mfi(symbol=comp,interval='1min', time_period=10)
main_data[comp]['MFI'] = mfi_data
# Fetch the 1-day rate of change of a triple smooth exponential moving average (TRIX) values
triX_train_data['AAPL'], meta_data = ti.get_trix(symbol=comp,interval='1min', time_period=10, series_type='close')
main_data[comp]['TRIX'] = triX_train_data['AAPL']
# Fetch the ultimate oscillator (ULTOSC) values
ultosc_data, meta_data = ti.get_ultsoc(symbol=comp,interval='1min', time_period=10)
main_data[comp]['ULTOSC'] = ultosc_data
# Fetch the directional movement index (DX) values
dX_train_data['AAPL'], meta_data = ti.get_dx(symbol=comp,interval='1min', time_period=10)
main_data[comp]['DX'] = dX_train_data['AAPL']
"""
# Fetch the Chaikin A/D line (AD) value
ad_data, meta_data = ti.get_trix(symbol=comp, interval='1min')
main_data[comp]['AD'] = ad_data
# Fetch the on balance volume (OBV) values
obv_data, meta_data = ti.get_obv(symbol=comp, interval='1min')
main_data[comp]['OBV'] = obv_data
# print(main_data[comp].head())
ts = TimeSeries(key='L2O2TYTG382ETN0N', output_format='pandas')
intraday_data, meta_data = ts.get_intraday(symbol=comp,
interval='1min',
outputsize='full')
# intraday_data = intraday_data.iloc[9:]
# intraday_data = intraday_data.reset_index(drop=True)
# intraday_data.index = main_data[comp].index
# intraday_data.set_index('date')
intraday_data.index = pd.Index(
[index[:-3] for index in intraday_data.index], name='date')
# intraday_data.set_index('date')
"""
for index in intraday_data.index:
print(index)
print(type(intraday_data.index))
"""
main_data[comp] = pd.concat([main_data[comp], intraday_data], axis=1)
print(main_data[comp].index)
print(main_data[comp].head())
main_data[comp] = main_data[comp].dropna()
main_data[comp].index.name = 'date'
y = np.where(
main_data[comp]['4. close'].shift(-1) >
main_data[comp]['4. close'], 1, -1)
main_data[comp]['Open-Close'] = main_data[comp]['1. open'] - main_data[
comp]['4. close']
main_data[comp]['High-Low'] = main_data[comp]['2. high'] - main_data[
comp]['3. low']
X = main_data[comp][main_data[comp].columns[0:]]
split = int(split_percentage * len(main_data[comp]['1. open']))
X_split_data[comp] = main_data[comp][split:]
X_train_data[comp] = X[:split]
y_train_data[comp] = y[:split]
X_test_data[comp] = X[split:]
y_test_data[comp] = y[split:]
return main_data
def technicalIndicators(API_key,ticker):
from alpha_vantage.techindicators import TechIndicators
import matplotlib.pyplot as plt
import mplfinance as mpf
ti=TechIndicators(key=API_key, output_format='pandas')
option=input('1. SMA\n2. EMA\n3. VWAP\n4. MACD\n5. STOCH\n6. RSI\n7. ADX\n8. CCI\n9. AROON\n10. BBANDS\n11. AD\n12. OBV\n').lower()
if option=='sma' or option=='1':
interval=int(input("Enter the interval:\n\t1. 1 minute\n\t2. 5 minutes\n\t3. 15 minutes\n\t4. 30 minutes\n\t5. 60 minutes\n\t6. daily \n\t7. weekly\n\t8. monthly\n"))
intervalList=['','1min','5min','15min','30min','60min','daily','weekly','monthly']
timeperiod=int(input('Enter Time Period\n'))
ser=int(input('Enter the series:\n\t1. close\n\t2. open\n\t3. low\n\t4. high\n'))
series=['','close','open','low','high']
data=ti.get_sma(symbol=ticker, interval=intervalList[interval], time_period=timeperiod, series_type=series[ser])[0]
data.columns = ['SMA']
data.index.name ="Date"
mpf.plot(data,
type='candle',
title=f'SMA for the {ticker} stock',
mav=(20),
volume=True,
tight_layout=True,
style='yahoo')
mpf.plot(data,type='line', volume=True)
return data
elif option=='ema' or option=='2':
interval=int(input("Enter the interval:\n\t1. 1 minute\n\t2. 5 minutes\n\t3. 15 minutes\n\t4. 30 minutes\n\t5. 60 minutes\n\t6. daily \n\t7. weekly\n\t8. monthly\n"))
intervalList=['','1min','5min','15min','30min','60min','daily','weekly','monthly']
timeperiod=int(input('Enter Time Period\n'))
ser=int(input('Enter the series:\n\t1. close\n\t2. open\n\t3. low\n\t4. high\n'))
series=['','close','open','low','high']
data=ti.get_ema(symbol=ticker, interval=intervalList[interval], time_period=timeperiod, series_type=series[ser])[0]
data.plot()
plt.title(f'EMA for the {ticker} stock')
plt.tight_layout()
plt.grid()
plt.show()
return data
elif option=='vwap' or option=='3':
interval=int(input("Enter the interval:\n\t1. 1 minute\n\t2. 5 minutes\n\t3. 15 minutes\n\t4. 30 minutes\n\t5. 60 minutes\n"))
intervalList=['','1min','5min','15min','30min','60min']
data=ti.get_vwap(symbol=ticker, interval=intervalList[interval])[0]
data.plot()
plt.title(f'VWAP for the {ticker} stock')
plt.tight_layout()
plt.grid()
plt.show()
return data
elif option=='macd' or option=='4':
interval=int(input("Enter the interval:\n\t1. 1 minute\n\t2. 5 minutes\n\t3. 15 minutes\n\t4. 30 minutes\n\t5. 60 minutes\n\t6. daily \n\t7. weekly\n\t8. monthly\n"))
intervalList=['','1min','5min','15min','30min','60min','daily','weekly','monthly']
ser=int(input('Enter the series:\n\t1. close\n\t2. open\n\t3. low\n\t4. high\n'))
series=['','close','open','low','high']
data=ti.get_macd(symbol=ticker, interval=intervalList[interval], series_type=series[ser])[0]
data.plot()
plt.title(f'MACD for the {ticker} stock')
plt.tight_layout()
plt.grid()
plt.show()
return data
elif option=='stoch' or option=='5':
interval=int(input("Enter the interval:\n\t1. 1 minute\n\t2. 5 minutes\n\t3. 15 minutes\n\t4. 30 minutes\n\t5. 60 minutes\n\t6. daily \n\t7. weekly\n\t8. monthly\n"))
intervalList=['','1min','5min','15min','30min','60min','daily','weekly','monthly']
data=ti.get_stoch(symbol=ticker, interval=intervalList[interval])[0]
data.plot()
plt.title(f'STOCH for the {ticker} stock')
plt.tight_layout()
plt.grid()
plt.show()
return data
elif option=='rsi' or option=='6':
interval=int(input("Enter the interval:\n\t1. 1 minute\n\t2. 5 minutes\n\t3. 15 minutes\n\t4. 30 minutes\n\t5. 60 minutes\n\t6. daily \n\t7. weekly\n\t8. monthly\n"))
intervalList=['','1min','5min','15min','30min','60min','daily','weekly','monthly']
timeperiod=int(input('Enter Time Period\n'))
ser=int(input('Enter the series:\n\t1. close\n\t2. open\n\t3. low\n\t4. high\n'))
series=['','close','open','low','high']
data=ti.get_rsi(symbol=ticker, interval=intervalList[interval], time_period=timeperiod, series_type=series[ser])[0]
data.plot()
plt.title(f'RSI for the {ticker} stock')
plt.tight_layout()
plt.grid()
plt.show()
return data
elif option=='adx' or option=='7':
interval=int(input("Enter the interval:\n\t1. 1 minute\n\t2. 5 minutes\n\t3. 15 minutes\n\t4. 30 minutes\n\t5. 60 minutes\n\t6. daily \n\t7. weekly\n\t8. monthly\n"))
intervalList=['','1min','5min','15min','30min','60min','daily','weekly','monthly']
timeperiod=int(input('Enter Time Period\n'))
data=ti.get_adx(symbol=ticker, interval=intervalList[interval], time_period=timeperiod)[0]
data.plot()
plt.title(f'ADX for the {ticker} stock')
plt.tight_layout()
plt.grid()
plt.show()
return data
elif option=='cci' or option=='8':
interval=int(input("Enter the interval:\n\t1. 1 minute\n\t2. 5 minutes\n\t3. 15 minutes\n\t4. 30 minutes\n\t5. 60 minutes\n\t6. daily \n\t7. weekly\n\t8. monthly\n"))
intervalList=['','1min','5min','15min','30min','60min','daily','weekly','monthly']
timeperiod=int(input('Enter Time Period\n'))
data=ti.get_cci(symbol=ticker, interval=intervalList[interval], time_period=timeperiod)[0]
data.plot()
plt.title(f'CCI for the {ticker} stock')
plt.tight_layout()
plt.grid()
plt.show()
return data
elif option=='aroon' or option=='9':
interval=int(input("Enter the interval:\n\t1. 1 minute\n\t2. 5 minutes\n\t3. 15 minutes\n\t4. 30 minutes\n\t5. 60 minutes\n\t6. daily \n\t7. weekly\n\t8. monthly\n"))
intervalList=['','1min','5min','15min','30min','60min','daily','weekly','monthly']
timeperiod=int(input('Enter Time Period'))
data=ti.get_aroon(symbol=ticker, interval=intervalList[interval], time_period=timeperiod)[0]
data.plot()
plt.title(f'AROON for the {ticker} stock')
plt.tight_layout()
plt.grid()
plt.show()
return data
elif option=='bbands' or option=='10':
interval=int(input("Enter the interval:\n\t1. 1 minute\n\t2. 5 minutes\n\t3. 15 minutes\n\t4. 30 minutes\n\t5. 60 minutes\n\t6. daily \n\t7. weekly\n\t8. monthly\n"))
intervalList=['','1min','5min','15min','30min','60min','daily','weekly','monthly']
timeperiod=int(input('Enter Time Period\n'))
ser=int(input('Enter the series:\n\t1. close\n\t2. open\n\t3. low\n\t4. high\n'))
series=['','close','open','low','high']
data=ti.get_bbands(symbol=ticker, interval=intervalList[interval], time_period=timeperiod, series_type=series[ser])[0]
data.plot()
plt.title(f'BBANDS for the {ticker} stock')
plt.tight_layout()
plt.grid()
plt.show()
return data
elif option=='ad' or option=='11':
interval=int(input("Enter the interval:\n\t1. 1 minute\n\t2. 5 minutes\n\t3. 15 minutes\n\t4. 30 minutes\n\t5. 60 minutes\n\t6. daily \n\t7. weekly\n\t8. monthly\n"))
intervalList=['','1min','5min','15min','30min','60min','daily','weekly','monthly']
data=ti.get_ad(symbol=ticker, interval=intervalList[interval])[0]
data.plot()
plt.title(f'AD for the {ticker} stock')
plt.tight_layout()
plt.grid()
plt.show()
return data
elif option=='obv' or option=='12':
interval=int(input("Enter the interval:\n\t1. 1 minute\n\t2. 5 minutes\n\t3. 15 minutes\n\t4. 30 minutes\n\t5. 60 minutes\n\t6. daily \n\t7. weekly\n\t8. monthly\n"))
intervalList=['','1min','5min','15min','30min','60min','daily','weekly','monthly']
data=ti.get_obv(symbol=ticker, interval=intervalList[interval])[0]
data.plot()
plt.title(f'OBV for the {ticker} stock')
plt.tight_layout()
plt.grid()
plt.show()
return data
else:
print("CANNOT RECOGNIZE")
def init_data(comp, interval):
# Input all the technical indicators and the stock times series data
# Initialize the AlphaVantage API call functions
ts = TimeSeries(key='L2O2TYTG382ETN0N', output_format='pandas')
ti = TechIndicators(key='L2O2TYTG382ETN0N', output_format='pandas')
# Other keys: QB5APLD84E50TQAC, L2O2TYTG382ETN0N
# Fetch the simple moving average (SMA) values
main_data, meta_data = ti.get_sma(symbol=comp,
interval='60min',
time_period=20,
series_type='close')
time.sleep(3)
# Fetch the exponential moving average (EMA) values
ema_data, meta_data = ti.get_ema(symbol=comp,
interval='60min',
time_period=20,
series_type='close')
main_data['EMA'] = ema_data
time.sleep(3)
"""
# Fetch the weighted moving average (WMA) values
wma_data, meta_data = ti.get_wma(symbol=comp,interval='60min', time_period=10, series_type='close')
main_data['WMA'] = wma_data
# Fetch the double exponential moving agerage (DEMA) values
dema_data, meta_data = ti.get_dema(symbol=comp,interval='60min', time_period=10, series_type='close')
main_data['DEMA'] = dema_data
# Fetch the triple exponential moving average (TEMA) values
tema_data, meta_data = ti.get_tema(symbol=comp,interval='60min', time_period=10, series_type='close')
main_data['TEMA'] = tema_data
# Fetch the triangular moving average (TRIMA) values
trima_data, meta_data = ti.get_trima(symbol=comp,interval='60min', time_period=10, series_type='close')
main_data['TRIMA'] = trima_data
# Fetch the Kaufman adaptive moving average (KAMA) values
kama_data, meta_data = ti.get_kama(symbol=comp,interval='60min', time_period=10, series_type='close')
main_data['KAMA'] = kama_data
# Fetch the MESA adaptive moving average (MAMA) values
mama_data, meta_data = ti.get_mama(symbol=comp,interval='60min', time_period=10, series_type='close')
main_data['MAMA'] = mama_data['MAMA']
main_data['FAMA'] = mama_data['FAMA']
# Fetch the triple exponential moving average (T3) values
t3_data, meta_data = ti.get_t3(symbol=comp,interval='60min', time_period=10, series_type='close')
main_data['T3'] = t3_data
"""
# Fetch the moving average convergence / divergence (MACD) values
macd_data, meta_data = ti.get_macd(symbol=comp,
interval='60min',
series_type='close')
main_data['MACD'] = macd_data['MACD']
main_data['MACD_Hist'] = macd_data['MACD_Hist']
main_data['MACD_Signal'] = macd_data['MACD_Signal']
time.sleep(3)
"""
# Fetch the moving average convergence / divergence values with controllable moving average type
macdext_data, meta_data = ti.get_macdext(symbol=comp,interval='60min', series_type='close')
main_data['MACDEXT'] = macdext_data['MACD']
main_data['MACDEXT_Hist'] = macdext_data['MACD_Hist']
main_data['MACDEXT_Signal'] = macdext_data['MACD_Signal']
"""
# Fetch the stochastic oscillator (STOCH) values
stoch_data, meta_data = ti.get_stoch(symbol=comp, interval='60min')
main_data['SlowK'] = stoch_data['SlowK']
main_data['SlowD'] = stoch_data['SlowD']
time.sleep(3)
"""
# Fetch the stochastic fast (STOCHF) values
stochf_data, meta_data = ti.get_stochf(symbol=comp,interval='60min')
main_data['FastK'] = stochf_data['FastK']
main_data['FastD'] = stochf_data['FastD']
"""
# Fetch the relative strength index (RSI) values
rsi_data, meta_data = ti.get_rsi(symbol=comp,
interval='60min',
time_period=10,
series_type='close')
main_data['RSI'] = rsi_data
time.sleep(3)
"""
# Fetch the stochastic relative strength index (STOCHRSI) values
stochrsi_data, meta_data = ti.get_stochrsi(symbol=comp,interval='60min', time_period=10, series_type='close')
main_data['STOCHRSI_FastK'] = stochrsi_data['FastK']
main_data['STOCHRSI_FastD'] = stochrsi_data['FastD']
# Fetch the Williams' %R (WILLR) values
willr_data, meta_data = ti.get_willr(symbol=comp,interval='60min', time_period=10)
main_data['WILLR'] = willr_data
"""
# Fetch the average directional movement index (ADX) values
adx_data, meta_data = ti.get_adx(symbol=comp,
interval='60min',
time_period=20)
main_data['ADX'] = adx_data
time.sleep(3)
"""
# Fetch the average directional movement index rating (ADXR) values
adxr_data, meta_data = ti.get_adxr(symbol=comp,interval='60min', time_period=10)
main_data['ADXR'] = adxr_data
# Fetch the absolute price oscillator (APO) values
apo_data, meta_data = ti.get_apo(symbol=comp,interval='60min', series_type='close')
main_data['APO'] = apo_data
# Fetch the percentage price oscillator (PPO) values
ppo_data, meta_data = ti.get_ppo(symbol=comp,interval='60min', series_type='close')
main_data['PPO'] = ppo_data
# Fetch the momentum (MOM) values
mom_data, meta_data = ti.get_mom(symbol=comp,interval='60min', time_period=10, series_type='close')
main_data['MOM'] = mom_data
# Fetch the balance of power (BOP) values
bop_data, meta_data = ti.get_bop(symbol=comp,interval='60min')
main_data['BOP'] = bop_data
"""
# Fetch the commodity channel index (CCI) values
cci_data, meta_data = ti.get_cci(symbol=comp,
interval='60min',
time_period=20)
main_data['CCI'] = cci_data
time.sleep(3)
"""
# Fetch the Chande momentum oscillator (CMO) values
cmo_data, meta_data = ti.get_cmo(symbol=comp,interval='60min', time_period=10, series_type='close')
main_data['CMO'] = cmo_data
# Fetch the rate of change (ROC) values
roc_data, meta_data = ti.get_roc(symbol=comp,interval='60min', time_period=10, series_type='close')
main_data['ROC'] = roc_data
# Fetch the rate of change ratio (ROCR) values
rocr_data, meta_data = ti.get_rocr(symbol=comp,interval='60min', time_period=10, series_type='close')
main_data['ROCR'] = rocr_data
time.sleep(5)
"""
# Fetch the Aroon (AROON) values
aroon_data, meta_data = ti.get_aroon(symbol=comp,
interval='60min',
time_period=20)
main_data['Aroon Down'] = aroon_data['Aroon Down']
main_data['Aroon Up'] = aroon_data['Aroon Up']
time.sleep(3)
"""
# Fetch the Aroon oscillator (AROONOSC) values
aroonosc_data, meta_data = ti.get_aroonosc(symbol=comp,interval='60min', time_period=10)
main_data['AROONOSC'] = aroonosc_data
# Fetch the money flow index (MFI) values
mfi_data, meta_data = ti.get_mfi(symbol=comp,interval='60min', time_period=10)
main_data['MFI'] = mfi_data
# Fetch the 1-day rate of change of a triple smooth exponential moving average (TRIX) values
triX_train_data['AAPL'], meta_data = ti.get_trix(symbol=comp,interval='60min', time_period=10, series_type='close')
main_data['TRIX'] = triX_train_data['AAPL']
# Fetch the ultimate oscillator (ULTOSC) values
ultosc_data, meta_data = ti.get_ultsoc(symbol=comp,interval='60min', time_period=10)
main_data['ULTOSC'] = ultosc_data
# Fetch the directional movement index (DX) values
dX_train_data['AAPL'], meta_data = ti.get_dx(symbol=comp,interval='60min', time_period=10)
main_data['DX'] = dX_train_data['AAPL']
"""
"""
# Fetch the Chaikin A/D line (AD) value
ad_data, meta_data = ti.get_ad(symbol=comp,interval='60min')
main_data['AD'] = ad_data
"""
# Fetch the on balance volume (OBV) values
obv_data, meta_data = ti.get_obv(symbol=comp, interval='60min')
main_data['OBV'] = obv_data
intraday_data, meta_data = ts.get_intraday(symbol=comp,
interval='60min',
outputsize='full')
intraday_data.index = pd.Index(
[index[:-3] for index in intraday_data.index], name='date')
#intraday_data.set_index('date')
main_data = pd.concat([main_data, intraday_data], axis=1)
print(main_data.index)
print(main_data.head())
main_data = main_data.dropna()
main_data.index.name = 'date'
company = comp
main_data.to_csv(f'./rsc/{company}_AV_data.csv', sep=',')
time.sleep(1)
print(comp)
time.sleep(5)
def save_ti(ticker, interval='daily'):
ti = TechIndicators(key='WS8K5VM8FZL52BTH', output_format='pandas')
ti.get_sma(
ticker,
interval)[0].to_csv(f'./Stock Data/TI/{ticker}_{interval}_sma.csv')
call_check()
ti.get_ema(
ticker,
interval)[0].to_csv(f'./Stock Data/TI/{ticker}_{interval}_ema.csv')
call_check()
ti.get_wma(
ticker,
interval)[0].to_csv(f'./Stock Data/TI/{ticker}_{interval}_wma.csv')
call_check()
ti.get_dema(
ticker,
interval)[0].to_csv(f'./Stock Data/TI/{ticker}_{interval}_dema.csv')
call_check()
ti.get_tema(
ticker,
interval)[0].to_csv(f'./Stock Data/TI/{ticker}_{interval}_tema.csv')
call_check()
ti.get_trima(
ticker,
interval)[0].to_csv(f'./Stock Data/TI/{ticker}_{interval}_trima.csv')
call_check()
ti.get_kama(
ticker,
interval)[0].to_csv(f'./Stock Data/TI/{ticker}_{interval}_kama.csv')
call_check()
ti.get_mama(
ticker,
interval)[0].to_csv(f'./Stock Data/TI/{ticker}_{interval}_mama.csv')
call_check()
ti.get_t3(
ticker,
interval)[0].to_csv(f'./Stock Data/TI/{ticker}_{interval}_t3.csv')
call_check()
ti.get_macd(
ticker,
interval)[0].to_csv(f'./Stock Data/TI/{ticker}_{interval}_macd.csv')
call_check()
ti.get_macdext(
ticker,
interval)[0].to_csv(f'./Stock Data/TI/{ticker}_{interval}_macdext.csv')
call_check()
ti.get_stoch(
ticker,
interval)[0].to_csv(f'./Stock Data/TI/{ticker}_{interval}_stoch.csv')
call_check()
ti.get_stochf(
ticker,
interval)[0].to_csv(f'./Stock Data/TI/{ticker}_{interval}_stochf.csv')
call_check()
ti.get_rsi(
ticker,
interval)[0].to_csv(f'./Stock Data/TI/{ticker}_{interval}_rsi.csv')
call_check()
ti.get_stochrsi(ticker, interval)[0].to_csv(
f'./Stock Data/TI/{ticker}_{interval}_stochris.csv')
call_check()
ti.get_willr(
ticker,
interval)[0].to_csv(f'./Stock Data/TI/{ticker}_{interval}_willr.csv')
call_check()
ti.get_adx(
ticker,
interval)[0].to_csv(f'./Stock Data/TI/{ticker}_{interval}_adx.csv')
call_check()
ti.get_adxr(
ticker,
interval)[0].to_csv(f'./Stock Data/TI/{ticker}_{interval}_adxr.csv')
call_check()
ti.get_apo(
ticker,
interval)[0].to_csv(f'./Stock Data/TI/{ticker}_{interval}_apo.csv')
call_check()
ti.get_ppo(
ticker,
interval)[0].to_csv(f'./Stock Data/TI/{ticker}_{interval}_ppo.csv')
call_check()
ti.get_mom(
ticker,
interval)[0].to_csv(f'./Stock Data/TI/{ticker}_{interval}_mom.csv')
call_check()
ti.get_bop(
ticker,
interval)[0].to_csv(f'./Stock Data/TI/{ticker}_{interval}_bop.csv')
call_check()
ti.get_cci(
ticker,
interval)[0].to_csv(f'./Stock Data/TI/{ticker}_{interval}_cci.csv')
call_check()
ti.get_cmo(
ticker,
interval)[0].to_csv(f'./Stock Data/TI/{ticker}_{interval}_cmo.csv')
call_check()
ti.get_roc(
ticker,
interval)[0].to_csv(f'./Stock Data/TI/{ticker}_{interval}_roc.csv')
call_check()
ti.get_rocr(
ticker,
interval)[0].to_csv(f'./Stock Data/TI/{ticker}_{interval}_rocr.csv')
call_check()
ti.get_aroon(
ticker,
interval)[0].to_csv(f'./Stock Data/TI/{ticker}_{interval}_aroon.csv')
call_check()
ti.get_aroonosc(ticker, interval)[0].to_csv(
f'./Stock Data/TI/{ticker}_{interval}_aroonosc.csv')
call_check()
ti.get_mfi(
ticker,
interval)[0].to_csv(f'./Stock Data/TI/{ticker}_{interval}_mfi.csv')
call_check()
ti.get_trix(
ticker,
interval)[0].to_csv(f'./Stock Data/TI/{ticker}_{interval}_trix.csv')
call_check()
ti.get_ultosc(
ticker,
interval)[0].to_csv(f'./Stock Data/TI/{ticker}_{interval}_ultosc.csv')
call_check()
ti.get_dx(
ticker,
interval)[0].to_csv(f'./Stock Data/TI/{ticker}_{interval}_dx.csv')
call_check()
ti.get_minus_di(ticker, interval)[0].to_csv(
f'./Stock Data/TI/{ticker}_{interval}_minus_di.csv')
call_check()
ti.get_plus_di(
ticker,
interval)[0].to_csv(f'./Stock Data/TI/{ticker}_{interval}_plus_di.csv')
call_check()
ti.get_minus_dm(ticker, interval)[0].to_csv(
f'./Stock Data/TI/{ticker}_{interval}_minus_dm.csv')
call_check()
ti.get_plus_dm(
ticker,
interval)[0].to_csv(f'./Stock Data/TI/{ticker}_{interval}_plus_dm.csv')
call_check()
ti.get_bbands(
ticker,
interval)[0].to_csv(f'./Stock Data/TI/{ticker}_{interval}_bbands.csv')
call_check()
ti.get_midpoint(ticker, interval)[0].to_csv(
f'./Stock Data/TI/{ticker}_{interval}_midpoint.csv')
call_check()
ti.get_midprice(ticker, interval)[0].to_csv(
f'./Stock Data/TI/{ticker}_{interval}_midprice.csv')
call_check()
ti.get_sar(
ticker,
interval)[0].to_csv(f'./Stock Data/TI/{ticker}_{interval}_sar.csv')
call_check()
ti.get_trange(
ticker,
interval)[0].to_csv(f'./Stock Data/TI/{ticker}_{interval}_trange.csv')
call_check()
ti.get_atr(
ticker,
interval)[0].to_csv(f'./Stock Data/TI/{ticker}_{interval}_atr.csv')
call_check()
ti.get_natr(
ticker,
interval)[0].to_csv(f'./Stock Data/TI/{ticker}_{interval}_natr.csv')
call_check()
ti.get_ad(
ticker,
interval)[0].to_csv(f'./Stock Data/TI/{ticker}_{interval}_ad.csv')
call_check()
ti.get_adosc(
ticker,
interval)[0].to_csv(f'./Stock Data/TI/{ticker}_{interval}_adosc.csv')
call_check()
ti.get_obv(
ticker,
interval)[0].to_csv(f'./Stock Data/TI/{ticker}_{interval}_obv.csv')
call_check()
ti.get_ht_trendline(ticker, interval)[0].to_csv(
f'./Stock Data/TI/{ticker}_{interval}_ht_trendline.csv')
call_check()
ti.get_ht_sine(
ticker,
interval)[0].to_csv(f'./Stock Data/TI/{ticker}_{interval}_ht_sine.csv')
call_check()
ti.get_ht_trendmode(ticker, interval)[0].to_csv(
f'./Stock Data/TI/{ticker}_{interval}_ht_trendmode.csv')
call_check()
ti.get_ht_dcperiod(ticker, interval)[0].to_csv(
f'./Stock Data/TI/{ticker}_{interval}_dcperiod.csv')
call_check()
ti.get_ht_dcphase(
ticker,
interval)[0].to_csv(f'./Stock Data/TI/{ticker}_{interval}_dcphase.csv')
call_check()
ti.get_ht_phasor(ticker, interval)[0].to_csv(
f'./Stock Data/TI/{ticker}_{interval}_ht_phasor.csv')
ti = TechIndicators(key='1HAYLUHGCB6E0VXC', output_format='pandas')
print("Progress: 0% complete...")
ad_data, meta_data = ti.get_ad(symbol=symbol, interval=interval)
adosc_data, meta_data = ti.get_adosc(symbol=symbol, interval=interval)
adx_data, meta_data = ti.get_adx(symbol=symbol, interval=interval)
adxr_data, meta_data = ti.get_adxr(symbol=symbol, interval=interval)
apo_data, meta_data = ti.get_apo(symbol=symbol, interval=interval)
aroon_data, meta_data = ti.get_aroon(symbol=symbol, interval=interval)
aroonosc_data, meta_data = ti.get_aroonosc(symbol=symbol, interval=interval)
atr_data, meta_data = ti.get_atr(symbol=symbol, interval=interval)
bop_data, meta_data = ti.get_bop(symbol=symbol, interval=interval)
cci_data, meta_data = ti.get_cci(symbol=symbol, interval=interval)
print("Progress: 20% complete...")
cmo_data, meta_data = ti.get_cmo(symbol=symbol, interval=interval)
dema_data, meta_data = ti.get_dema(symbol=symbol, interval=interval)
dx_data, meta_data = ti.get_dx(symbol=symbol, interval=interval)
ema_data, meta_data = ti.get_ema(symbol=symbol, interval=interval)
ht_dcperiod_data, meta_data = ti.get_ht_dcperiod(symbol=symbol, interval=interval)
ht_dcphase_data, meta_data = ti.get_ht_dcphase(symbol=symbol, interval=interval)
ht_phasor_data, meta_data = ti.get_ht_phasor(symbol=symbol, interval=interval)
ht_sine_data, meta_data = ti.get_ht_sine(symbol=symbol, interval=interval)
ht_trendline_data, meta_data = ti.get_ht_trendline(symbol=symbol, interval=interval)
ht_trendmode_data, meta_data = ti.get_ht_trendmode(symbol=symbol, interval=interval)
print("Progress: 40% complete...")
def preds(api_key):
df = pd.read_csv('GOOGL.csv', index_col='Date')
preds = pd.DataFrame()
for i in np.arange(start=1, stop=6):
if preds.empty:
preds = pd.DataFrame(data=df['close'].pct_change(1))
preds.columns = ['today']
else:
lags = pd.DataFrame(df['close'].pct_change(i))
lags.columns = ['lag{}'.format(i)]
preds = preds.join(lags)
preds = preds.join(df['volume'])
preds = preds.join(df['close'])
preds['dir'] = (preds['close'] > preds['close'].shift(1)).shift(-1)
filename = 'GOOGL-lags.csv'
try:
preds.to_csv('{}'.format(filename))
print('GOOGL lags saved successfully as "{}"'.format(filename))
except:
print('Unable to save "{}"'.format(filename))
ti = TechIndicators(key=api_key, output_format='pandas')
sma, _ = ti.get_sma(symbol='GOOGL', interval='daily', series_type='close')
ema, _ = ti.get_ema(symbol='GOOGL', interval='daily', series_type='close')
wma, _ = ti.get_wma(symbol='GOOGL', interval='daily', series_type='close')
trima, _ = ti.get_trima(symbol='GOOGL',
interval='daily',
series_type='close')
kama, _ = ti.get_kama(symbol='GOOGL',
interval='daily',
series_type='close')
macd, _ = ti.get_macd(symbol='GOOGL',
interval='daily',
series_type='close')
rsi, _ = ti.get_rsi(symbol='GOOGL', interval='daily', series_type='close')
adx, _ = ti.get_adx(symbol='GOOGL', interval='daily')
apo, _ = ti.get_apo(symbol='GOOGL', interval='daily', series_type='close')
stoch, _ = ti.get_stoch(symbol='GOOGL', interval='daily')
ppo, _ = ti.get_ppo(symbol='GOOGL', interval='daily')
mom, _ = ti.get_mom(symbol='GOOGL', interval='daily', series_type='close')
cci, _ = ti.get_cci(symbol='GOOGL', interval='daily')
aroon, _ = ti.get_aroon(symbol='GOOGL',
interval='daily',
series_type='close')
mfi, _ = ti.get_mfi(symbol='GOOGL', interval='daily', series_type='close')
dx, _ = ti.get_dx(symbol='GOOGL', interval='daily', series_type='close')
ad, _ = ti.get_ad(symbol='GOOGL', interval='daily')
obv, _ = ti.get_obv(symbol='GOOGL', interval='daily')
filename2 = 'GOOGL-techs.csv'
try:
preds_test = pd.concat([
sma, ema, wma, trima, kama, macd, rsi, adx, apo, stoch, ppo, mom,
cci, aroon, mfi, dx, ad, obv
],
axis=1)
pd.DataFrame(preds_test).to_csv('{}'.format(filename2))
print('Google techs saved as successfully as "{}"'.format(filename2))
except:
print('Unable to save "{}"'.format(filename2))
df['roc'], _ = ti.get_roc(symbol=symbol, time_period=roc_time_period)
df['sma'], _ = ti.get_sma(symbol=symbol, time_period=sma_time_period)
df['ema'], _ = ti.get_ema(symbol=symbol, time_period=ema_time_period)
df['rsi'], _ = ti.get_rsi(symbol=symbol, time_period=rsi_time_period)
df['obv'], _ = ti.get_obv(symbol=symbol)
macd = ti.get_macd(symbol=symbol)[0].sort_index(axis=0, ascending=True)
for column in macd.columns:
df[column] = macd[column]
stoch = ti.get_stoch(symbol=symbol)[0].sort_index(axis=0, ascending=True)
for column in stoch.columns:
df[column] = stoch[column]
df['adx'], _ = ti.get_adx(symbol=symbol, time_period=adx_time_period)
df['cci'], _ = ti.get_cci(symbol=symbol, time_period=cci_time_period)
aroon = ti.get_aroon(
symbol=symbol, time_period=aroon_time_period)[0].sort_index(axis=0,
ascending=True)
for column in aroon.columns:
df[column] = aroon[column]
bbands = ti.get_bbands(symbol=symbol,
time_period=bbands_time_period)[0].sort_index(
axis=0, ascending=True)
for column in bbands.columns:
df[column] = bbands[column]
df['ad'], _ = ti.get_ad(symbol=symbol)
df['shifted_ROC_BOOL'] = (df['roc'].shift(-1) >= 0).astype(int)
def techIndicatorPull(symbol):
interval = "1min"
print("Calculating Tech Indicators ...")
ti = TechIndicators(key='1HAYLUHGCB6E0VXC', output_format='pandas')
print("Progress: 0% complete...")
ad_data, meta_data = ti.get_ad(symbol=symbol, interval=interval)
adosc_data, meta_data = ti.get_adosc(symbol=symbol, interval=interval)
adx_data, meta_data = ti.get_adx(symbol=symbol, interval=interval)
adxr_data, meta_data = ti.get_adxr(symbol=symbol, interval=interval)
apo_data, meta_data = ti.get_apo(symbol=symbol, interval=interval)
aroon_data, meta_data = ti.get_aroon(symbol=symbol, interval=interval)
aroonosc_data, meta_data = ti.get_aroonosc(symbol=symbol,
interval=interval)
atr_data, meta_data = ti.get_atr(symbol=symbol, interval=interval)
bop_data, meta_data = ti.get_bop(symbol=symbol, interval=interval)
cci_data, meta_data = ti.get_cci(symbol=symbol, interval=interval)
print("Progress: 20% complete...")
cmo_data, meta_data = ti.get_cmo(symbol=symbol, interval=interval)
dema_data, meta_data = ti.get_dema(symbol=symbol, interval=interval)
dx_data, meta_data = ti.get_dx(symbol=symbol, interval=interval)
ema_data, meta_data = ti.get_ema(symbol=symbol, interval=interval)
ht_dcperiod_data, meta_data = ti.get_ht_dcperiod(symbol=symbol,
interval=interval)
ht_dcphase_data, meta_data = ti.get_ht_dcphase(symbol=symbol,
interval=interval)
ht_phasor_data, meta_data = ti.get_ht_phasor(symbol=symbol,
interval=interval)
ht_sine_data, meta_data = ti.get_ht_sine(symbol=symbol, interval=interval)
ht_trendline_data, meta_data = ti.get_ht_trendline(symbol=symbol,
interval=interval)
ht_trendmode_data, meta_data = ti.get_ht_trendmode(symbol=symbol,
interval=interval)
print("Progress: 40% complete...")
kama_data, meta_data = ti.get_kama(symbol=symbol, interval=interval)
macd_data, meta_data = ti.get_macd(symbol=symbol, interval=interval)
macdext_data, meta_data = ti.get_macdext(symbol=symbol, interval=interval)
mama_data, meta_data = ti.get_mama(symbol=symbol, interval=interval)
mfi_data, meta_data = ti.get_mfi(symbol=symbol, interval=interval)
midpoint_data, meta_data = ti.get_midpoint(symbol=symbol,
interval=interval)
midprice_data, meta_data = ti.get_midprice(symbol=symbol,
interval=interval)
minus_di_data, meta_data = ti.get_minus_di(symbol=symbol,
interval=interval)
mom_data, meta_data = ti.get_mom(symbol=symbol, interval=interval)
natr_data, meta_data = ti.get_natr(symbol=symbol, interval=interval)
print("Progress: 60% complete...")
obv_data, meta_data = ti.get_obv(symbol=symbol, interval=interval)
plus_di_data, meta_data = ti.get_plus_di(symbol=symbol, interval=interval)
plus_dm_data, meta_data = ti.get_plus_dm(symbol=symbol, interval=interval)
ppo_data, meta_data = ti.get_ppo(symbol=symbol, interval=interval)
roc_data, meta_data = ti.get_roc(symbol=symbol, interval=interval)
rocr_data, meta_data = ti.get_rocr(symbol=symbol, interval=interval)
rsi_data, meta_data = ti.get_rsi(symbol=symbol, interval=interval)
sar_data, meta_data = ti.get_sar(symbol=symbol, interval=interval)
sma_data, meta_data = ti.get_sma(symbol=symbol, interval=interval)
stoch_data, meta_data = ti.get_stoch(symbol=symbol, interval=interval)
print("Progress: 80% complete...")
stochf_data, meta_data = ti.get_stochf(symbol=symbol, interval=interval)
stochrsi_data, meta_data = ti.get_stochrsi(symbol=symbol,
interval=interval)
t3_data, meta_data = ti.get_t3(symbol=symbol, interval=interval)
tema_data, meta_data = ti.get_tema(symbol=symbol, interval=interval)
trange_data, meta_data = ti.get_trange(symbol=symbol, interval=interval)
trima_data, meta_data = ti.get_trima(symbol=symbol, interval=interval)
trix_data, meta_data = ti.get_trix(symbol=symbol, interval=interval)
ultsoc_data, meta_data = ti.get_ultsoc(symbol=symbol, interval=interval)
willr_data, meta_data = ti.get_willr(symbol=symbol, interval=interval)
wma_data, meta_data = ti.get_wma(symbol=symbol, interval=interval)
bbands_data, meta_data = ti.get_bbands(symbol=symbol, interval=interval)
print("Progress: 90% complete...")
result = pd.concat([
ad_data, adosc_data, adx_data, adxr_data, apo_data, aroon_data,
aroonosc_data, atr_data, bop_data, cci_data, cmo_data, dema_data,
dx_data, ema_data, ht_dcperiod_data, ht_dcphase_data, ht_phasor_data,
ht_sine_data, ht_trendline_data, ht_trendmode_data, kama_data,
macd_data, macdext_data, mama_data, mfi_data, midpoint_data,
midprice_data, minus_di_data, mom_data, natr_data, obv_data,
plus_di_data, plus_dm_data, ppo_data, roc_data, rocr_data, rsi_data,
sar_data, sma_data, stoch_data, stochf_data, stochrsi_data, t3_data,
tema_data, trange_data, trima_data, trix_data, ultsoc_data, willr_data,
wma_data, bbands_data
],
axis=1)
result["stockticker"] = symbol
result = result.reindex(result.index.rename('timestamp'))
result = result.dropna(axis=0, how='any')
result = result.round(2)
result = result.rename(
columns={
'Chaikin A/D': 'ChaikinAD',
'Aroon Up': 'AroonUp',
'Aroon Down': 'AroonDown',
'LEAD SINE': 'LeadSine',
'Real Upper Band': 'RealUpperBand',
'Real Middle Band': 'RealMiddleBand',
'Real Lower Band': 'RealLowerBand'
})
fd = open('results/rawTechIndicatorData.csv', 'w')
result.to_csv(fd, index=True, encoding='utf-8')
fd.close()
print("Progress: 99% complete...")
df = pd.read_csv('results/rawTechIndicatorData.csv')
insertToOracleDatabase(
df,
'''insert /*+ ignore_row_on_dupkey_index(stocktechindicator, stocktechindicator_pk) */
into stocktechindicator
(timestamp,ChaikinAD,ADOSC,ADX,ADXR,
APO,AroonUp,AroonDown,AROONOSC,ATR,
BOP,CCI,CMO,DEMA,DX,
EMA,DCPERIOD,HT_DCPHASE,QUADRATURE, PHASE,
LEADSINE,SINE,HT_TRENDLINE,TRENDMODE,KAMA,
MACD,MACD_Hist,MACD_Signal,MACD_Signal2,MACD2,
MACD_Hist2,FAMA,MAMA,MFI,MIDPOINT,
MIDPRICE,MINUS_DI,MOM,NATR,OBV,
PLUS_DI,PLUS_DM,PPO,ROC,ROCR,
RSI,SAR,SMA,SlowK,SlowD,
FastK,FastD,FastK2,FastD2,T3,
TEMA,TRANGE,TRIMA,TRIX,ULTOSC,
WILLR,WMA,RealLowerBand,RealUpperBand,RealMiddleBand,
stockticker)
values (:1, :2, :3, :4, :5,
:6, :7, :8, :9, :10,
:11, :12, :13, :14, :15,
:16, :17, :18, :19, :20,
:21, :22, :23, :24, :25,
:26, :27, :28, :29, :30,
:31, :32, :33, :34, :35,
:36, :37, :38, :39, :40,
:41, :42, :43, :44, :45,
:46, :47, :48, :49, :50,
:51, :52, :53, :54, :55,
:56, :57, :58, :59, :60,
:61, :62, :63, :64, :65,
:66)''')
def getData(tker,alg,time,tDF,tI,incP,typeClass,return_list):
print("Getting Data")
dataDict= []
symb = tker
#Do not use 60min since that does not work.
#Cannot figure out how, but the keys for date and time are different.
tInterval = time
ourKey = '2ONP54KRQ2G73D3K'
#Using the wrapper class to get the different data for the inputted ticker
#In a try-catch because sometime the API fails
try:
#Making the retry number high so the API call should not fail
ts = TimeSeries(key = ourKey, output_format = 'json',retries=10000)
#tInterval has first letter capitalized when getting input for neatness, it is then lowered here for the API call
#Here we are getting the price data for the requested time interval
if tInterval == 'Daily':
priceData, meta_priceData = ts.get_daily(symbol= symb, outputsize='full')
tInterval = 'daily'
elif tInterval == 'Weekly':
priceData, meta_priceData = ts.get_weekly(symbol = symb)
tInterval = 'weekly'
elif tInterval =='Monthly':
priceData, meta_priceData = ts.get_monthly(symbol = symb)
tInterval = 'monthly'
else:
priceData, meta_priceData = ts.get_intraday(symbol = symb, interval = tInterval, outputsize='full')
#Intraday date and time key includes seconds, while it does not for technicial indicators
for day in priceData.copy():
priceData[day[:-3]] = priceData.pop(day)
ti = TechIndicators(key = ourKey, output_format ='json')
#Here we get the technical indicators the user chose from the menu, unless they chose to do a tDF.
if tDF == "No":
if 0 in tI:
macD, meta_macD = ti.get_macd(symbol = symb, interval = tInterval, series_type = 'close')
dataDict.append(macD)
if 1 in tI:
stoch, meta_stoch = ti.get_stoch(symbol = symb, interval = tInterval, fastkperiod = '5')
dataDict.append(stoch)
if 2 in tI:
ema, meta_ema = ti.get_ema(symbol = symb, interval = tInterval, time_period = '8', series_type = 'close')
dataDict.append(ema)
if 3 in tI:
rsi, meta_rsi = ti.get_rsi(symbol = symb, interval = tInterval, time_period = '14', series_type='close')
dataDict.append(rsi)
if 4 in tI:
plusDI, meta_plusDI = ti.get_plus_di(symbol = symb, interval = tInterval, time_period = '14')
minusDI, meta_minusDI = ti.get_minus_di(symbol = symb, interval = tInterval, time_period = '14')
dataDict.append(plusDI)
dataDict.append(minusDI)
if 5 in tI:
cci, meta_cci = ti.get_cci(symbol = symb, interval = tInterval, time_period = '14')
dataDict.append(cci)
if 6 in tI:
willR, meta_willR = ti.get_willr(symbol = symb, interval = tInterval, time_period = '14')
dataDict.append(willR)
if 7 in tI:
oBV, meta_OBV = ti.get_obv(symbol = symb, interval = tInterval)
dataDict.append(oBV)
if 8 in tI:
aD, meta_AD = ti.get_ad(symbol = symb, interval = tInterval)
dataDict.append(aD)
if 9 in tI:
bBands, meta_bBands = ti.get_bbands(symbol = symb, interval = tInterval, time_period = '14', series_type = 'close')
dataDict.append(bBands)
if 10 in tI:
aroon, meta_aroon = ti.get_aroon(symbol = symb, interval = tInterval, time_period = '14')
dataDict.append(aroon)
if 11 in tI:
adx, meta_adx = ti.get_adx(symbol = symb, interval = tInterval, time_period = '14')
dataDict.append(adx)
if(incP == "Yes"):
dataDict.append(priceData)
#Since a tDF requires a manual transformation, we choose the technical indicators for it, which are retrieved here
else:
macD, meta_macD = ti.get_macd(symbol = symb, interval = tInterval, series_type = 'close')
stoch, meta_stoch = ti.get_stoch(symbol = symb, interval = tInterval, fastkperiod = '5')
ema, meta_ema = ti.get_ema(symbol = symb, interval = tInterval, time_period = '8', series_type = 'close')
rsi, meta_rsi = ti.get_rsi(symbol = symb, interval = tInterval, time_period = '14', series_type='close')
plusDI, meta_plusDI = ti.get_plus_di(symbol = symb, interval = tInterval, time_period = '14')
minusDI, meta_minusDI = ti.get_minus_di(symbol = symb, interval = tInterval, time_period = '14')
cci, meta_cci = ti.get_cci(symbol = symb, interval = tInterval, time_period = '14')
willR, meta_willR = ti.get_willr(symbol = symb, interval = tInterval, time_period = '14')
dataDict.append(macD)
dataDict.append(stoch)
dataDict.append(ema)
dataDict.append(rsi)
dataDict.append(plusDI)
dataDict.append(minusDI)
dataDict.append(cci)
dataDict.append(willR)
#Returning with nothing tells our main process that there was an error
except ValueError:
return
#Sometimes the API may return a empty list which will cause an error for the training
for list in dataDict:
if not list:
return
#Need atleast one of the dictionaries to be ordered by date, so we can organize all our data.
priceData = collections.OrderedDict(sorted(priceData.items()))
#dates used for drawing graph afterwards
dates = []
for day in priceData:
dates.append(day)
#Parsing the data
#All the data for each day is put into an array, which is put into trainingData, sorted by dates
trainingData = []
#Putting the ordered price data into priceList, instead of using a Dictionary
#So if the user doesn't want to include price in training, we still have a list to train
#for the actual outputs/
priceList = []
#If the user is using this in the middle of the day/week/month, the data will have a partial time interval up to the current time
#Here we remove the last day if it is the same date as today, so we can get a guess for the timer interval asked.
todaysDate = str(datetime.datetime.now())[:10]
if todaysDate in priceData.keys():
del priceData[todaysDate]
for day in priceData:
#Making sure the day is in the priceData and each of the technical indicator lists.
#The API doesn't necessarily give the data for every data for the different requests
if all(day in indic for indic in dataDict):
vector = []
priceData[day] = collections.OrderedDict(sorted(priceData[day].items()))
for stat in priceData[day]:
vector.append(float(priceData[day][stat]))
priceList.append(vector)
vector = []
for x in range(0,len(dataDict)):
#Order the different stats within a data, so the order is consistent across vectors
dataDict[x][day] = collections.OrderedDict(sorted(dataDict[x][day].items()))
#Combine all the different data into a vector
for x in range(0,len(dataDict)):
for stat in dataDict[x][day]:
vector.append(float(dataDict[x][day][stat]))
trainingData.append(vector)
#If the user wanted to do a tDF, the training data is transformed, otherwise the trainingData goes straight to PredictAndTest
trendDeter = False if tDF == "No" else True
#delete first element of priceList because trendDeter, remove first element form training data
if trendDeter:
del priceList[0]
predictAndTest(trendDeterTransform(trainingData), priceList, trendDeter, alg, tInterval, dates, typeClass, return_list)
else:
predictAndTest(trainingData, priceList, trendDeter, alg, tInterval, dates, typeClass, return_list)
class DataLoader:
"""
API limit: 5 API requests per minute and 500 requests per day
Get your free API key from
https://www.alphavantage.co/support/#support
FOREX, while supported, is not implemented.
API details here:
https://github.com/RomelTorres/alpha_vantage
"""
def __init__(self, api_key):
self.API_KEY = api_key
self.cc = CryptoCurrencies(key=self.API_KEY, output_format='pandas')
self.ti = TechIndicators(key=self.API_KEY, output_format='pandas')
self.ts = TimeSeries(key=self.API_KEY, output_format='pandas')
def get_local_data(self, path):
df = pd.read_csv(path)
return df.sort_values('Timestamp')
def get_crypto(self, crypto_symbol):
data, _ = self.cc.get_digital_currency_daily(symbol=crypto_symbol,
market='CNY')
return rename_data(data, crypto=True)
def get_stock(self,
stock: str,
period=TimeFrame.TimeFrame.DAILY,
full: bool = False):
"""
Returns stock data and meta data of the ticker for the specified time frame
:param stock: [ENUM]: Stock ticker
:param period: [ENUM] DAILY, WEEKLY, MONTHLY
:param full: Returns only last 100 ticks if False, otherwise full tick data set if True. False by default.
:return: stock data, meta_data
"""
if period is TimeFrame.TimeFrame.DAILY:
if full:
data, _ = self.ts.get_daily(symbol=stock.upper(),
outputsize='full')
return rename_data(data)
else:
data, _ = self.ts.get_daily(symbol=stock.upper(),
outputsize='compact')
return rename_data(data)
if period is TimeFrame.TimeFrame.WEEKLY:
if full:
data, _ = data = self.ts.get_weekly(symbol=stock.upper(),
outputsize='full')
return rename_data(data)
else:
data, _ = self.ts.get_weekly(symbol=stock.upper(),
outputsize='compact')
return rename_data(data)
if period is TimeFrame.TimeFrame.MONTHLY:
if full:
data, _ = self.ts.get_monthly(symbol=stock.upper(),
outputsize='full')
return rename_data(data)
else:
data, _ = self.ts.get_monthly(symbol=stock.upper(),
outputsize='compact')
return rename_data(data)
def get_intraday_stock(self,
stock: str,
interval: INTERVAL = INTERVAL.INTERVAL.FIVE_MIN,
full: bool = False):
"""
Returns intraday tick data for the given stock in the given time interval
:param stock: [ENUM] ticker
:param interval: [ENUM] 1min, 5min, 15min, 30min, 60min
:param full: Returns only last 100 ticks if False, otherwise full tick data set if True. False by default.
:return: stock data, meta_data
"""
if full:
data, _ = self.ts.get_intraday(symbol=stock.upper(),
interval=interval.value,
outputsize='full')
return rename_data(data)
else:
data, _ = self.ts.get_intraday(symbol=stock.upper(),
interval=interval.value,
outputsize='compact')
return rename_data(data)
def get_tech_indicator(
self,
stock="MSFT",
indicator=TECHIND.TECHIND.BBANDS,
interval=TimeFrame.TimeFrame.DAILY,
time_period: int = 20,
):
"""
Returns the technical indicator for the given stock ticker on the given interval
:param indicator [ENUM]: @See TECHIND
:param stock [ENUM]: Ticker
:param interval: Daily, Weekly, Monthly. Set to Daily by default
:param time_period: Nr of time units between two calculating points. Set to 20 by default.
:return: pandas dataframe containing the technical indicator for all recorded trading days of the stock.
"""
if indicator is TECHIND.TECHIND.BBANDS:
data, _ = self.ti.get_bbands(symbol=stock.upper(),
interval=interval.name.lower(),
time_period=time_period)
if indicator is TECHIND.TECHIND.SMA:
data, _ = self.ti.get_sma(symbol=stock.upper(),
interval=interval.name.lower(),
time_period=time_period)
if indicator is TECHIND.TECHIND.EMA:
data, _ = self.ti.get_ema(symbol=stock.upper(),
interval=interval.name.lower(),
time_period=time_period)
if indicator is TECHIND.TECHIND.WMA:
data, _ = self.ti.get_wma(symbol=stock.upper(),
interval=interval.name.lower(),
time_period=time_period)
if indicator is TECHIND.TECHIND.MACD:
data, _ = self.ti.get_macd(symbol=stock.upper(),
interval=interval.name.lower())
if indicator is TECHIND.TECHIND.STOCH:
data, _ = self.ti.get_stoch(symbol=stock.upper(),
interval=interval.name.lower())
if indicator is TECHIND.TECHIND.RSI:
data, _ = self.ti.get_rsi(symbol=stock.upper(),
interval=interval.name.lower(),
time_period=time_period)
if indicator is TECHIND.TECHIND.ADX:
data, _ = self.ti.get_adx(symbol=stock.upper(),
interval=interval.name.lower(),
time_period=time_period)
if indicator is TECHIND.TECHIND.CCI:
data, _ = self.ti.get_cci(symbol=stock.upper(),
interval=interval.name.lower(),
time_period=time_period)
if indicator is TECHIND.TECHIND.AROON:
data, _ = self.ti.get_aroon(symbol=stock.upper(),
interval=interval.name.lower(),
time_period=time_period)
if indicator is TECHIND.TECHIND.MOM:
data, _ = self.ti.get_mom(symbol=stock.upper(),
interval=interval.name.lower())
if indicator is TECHIND.TECHIND.OBV:
data, _ = self.ti.get_obv(symbol=stock.upper(),
interval=interval.name.lower())
return data
