|
|
|
|
|
import pandas as pd |
|
|
import talib |
|
|
import numpy as np |
|
|
|
|
|
def calculate_indicators(df): |
|
|
""" |
|
|
Calculate all possible indicators from TA-Lib for a given OHLCV DataFrame. |
|
|
df: DataFrame with columns ['Open','High','Low','Close','Volume'] |
|
|
Returns: dict of {indicator_name: Series or DataFrame} |
|
|
""" |
|
|
indicators = {} |
|
|
|
|
|
close = df['Close'] |
|
|
high = df['High'] |
|
|
low = df['Low'] |
|
|
open_ = df['Open'] |
|
|
volume = df['Volume'] |
|
|
|
|
|
|
|
|
indicators['SMA_5'] = talib.SMA(close, timeperiod=5) |
|
|
indicators['SMA_10'] = talib.SMA(close, timeperiod=10) |
|
|
indicators['SMA_20'] = talib.SMA(close, timeperiod=20) |
|
|
indicators['SMA_50'] = talib.SMA(close, timeperiod=50) |
|
|
indicators['EMA_5'] = talib.EMA(close, timeperiod=5) |
|
|
indicators['EMA_10'] = talib.EMA(close, timeperiod=10) |
|
|
indicators['EMA_20'] = talib.EMA(close, timeperiod=20) |
|
|
indicators['EMA_50'] = talib.EMA(close, timeperiod=50) |
|
|
|
|
|
|
|
|
indicators['ADX'] = talib.ADX(high, low, close, timeperiod=14) |
|
|
indicators['CCI'] = talib.CCI(high, low, close, timeperiod=14) |
|
|
indicators['AROON_UP'], indicators['AROON_DOWN'] = talib.AROON(high, low, timeperiod=14) |
|
|
indicators['MACD'], indicators['MACD_signal'], indicators['MACD_hist'] = talib.MACD(close) |
|
|
indicators['ATR'] = talib.ATR(high, low, close, timeperiod=14) |
|
|
|
|
|
|
|
|
indicators['RSI'] = talib.RSI(close, timeperiod=14) |
|
|
indicators['STOCH_slowk'], indicators['STOCH_slowd'] = talib.STOCH(high, low, close) |
|
|
indicators['STOCHF_fastk'], indicators['STOCHF_fastd'] = talib.STOCHF(high, low, close) |
|
|
indicators['WILLR'] = talib.WILLR(high, low, close, timeperiod=14) |
|
|
|
|
|
|
|
|
indicators['BB_upper'], indicators['BB_middle'], indicators['BB_lower'] = talib.BBANDS(close) |
|
|
indicators['ATR'] = talib.ATR(high, low, close, timeperiod=14) |
|
|
|
|
|
|
|
|
try: |
|
|
indicators['SuperTrend'] = supertrend(df) |
|
|
except Exception as e: |
|
|
indicators['SuperTrend'] = pd.Series([np.nan]*len(df), index=df.index) |
|
|
|
|
|
return indicators |
|
|
|
|
|
|
|
|
def supertrend(df, period=10, multiplier=3): |
|
|
""" |
|
|
Compute SuperTrend indicator. |
|
|
Returns a Series same length as df with SuperTrend values. |
|
|
""" |
|
|
high = df['High'] |
|
|
low = df['Low'] |
|
|
close = df['Close'] |
|
|
|
|
|
atr = talib.ATR(high, low, close, timeperiod=period) |
|
|
hl2 = (high + low) / 2 |
|
|
upperband = hl2 + multiplier * atr |
|
|
lowerband = hl2 - multiplier * atr |
|
|
|
|
|
supertrend = pd.Series(index=df.index) |
|
|
direction = True |
|
|
|
|
|
for i in range(len(df)): |
|
|
if i == 0: |
|
|
supertrend.iloc[i] = hl2.iloc[i] |
|
|
continue |
|
|
if close.iloc[i] > supertrend.iloc[i-1]: |
|
|
direction = True |
|
|
elif close.iloc[i] < supertrend.iloc[i-1]: |
|
|
direction = False |
|
|
|
|
|
if direction: |
|
|
supertrend.iloc[i] = lowerband.iloc[i] if lowerband.iloc[i] > supertrend.iloc[i-1] else supertrend.iloc[i-1] |
|
|
else: |
|
|
supertrend.iloc[i] = upperband.iloc[i] if upperband.iloc[i] < supertrend.iloc[i-1] else supertrend.iloc[i-1] |
|
|
|
|
|
return supertrend |
|
|
|
