def setUpClass(cls):
cls._df = pd.read_csv(cls._filename, sep=',')
cls._params = dict(high=cls._df['High'],
low=cls._df['Low'],
close=cls._df['Close'],
s=7,
m=14,
len=28,
ws=4.0,
wm=2.0,
wl=1.0,
fillna=False)
cls._indicator = UltimateOscillator(**cls._params)
def setUpClass(cls):
cls._df = pd.read_csv(cls._filename, sep=",")
cls._params = dict(
high=cls._df["High"],
low=cls._df["Low"],
close=cls._df["Close"],
window1=7,
window2=14,
window3=28,
weight1=4.0,
weight2=2.0,
weight3=1.0,
fillna=False,
)
cls._indicator = UltimateOscillator(**cls._params)
class TestUltimateOscillator(unittest.TestCase):
"""
https://school.stockcharts.com/doku.php?id=technical_indicators:ultimate_oscillator
"""
_filename = 'ta/tests/data/cs-ultosc.csv'
def setUp(self):
self._df = pd.read_csv(self._filename, sep=',')
self._indicator = UltimateOscillator(
high=self._df['High'], low=self._df['Low'], close=self._df['Close'],
s=7, m=14, len=28, ws=4.0, wm=2.0, wl=1.0, fillna=False)
def tearDown(self):
del(self._df)
def test_uo(self):
target = 'Ult_Osc'
result = self._indicator.uo()
pd.testing.assert_series_equal(self._df[target].tail(), result.tail(), check_names=False)
def get_stock_predictions():
start = time.time()
ticker = request.args.get('ticker')
date = -1
np.random.seed(312)
try:
df = stocks[ticker].copy()
df['return'] = df['PRC'].pct_change()
df.drop( columns = ['COMNAM'], inplace = True)
df.set_index('date', inplace=True, drop=True)
df['1_day_return'] = df['return'].shift(-1)
df['1_day_return'] = np.where(df['1_day_return'] > 0, 1, 0)
ewma = pd.Series.ewm
df['Fast_Exp_Moving_Avg'] = df['PRC'].transform(lambda x: ewma(x, span = 12).mean())
df['Slow_Exp_Moving_Avg'] = df['PRC'].transform(lambda x: ewma(x, span = 26).mean())
df['Momentum_Factor'] = df['Fast_Exp_Moving_Avg']/df['Slow_Exp_Moving_Avg']
df['Moving_Avg'] = df['PRC'].transform(lambda x: x.rolling(window=20).mean())
df['Moving_Std_Deviation'] = df['PRC'].transform(lambda x: x.rolling(window=20).std())
df['Ultimate_Oscillator'] = UltimateOscillator(df['ASKHI'], df['BIDLO'], df['PRC'], fillna = True).uo()
df['RSI_Indicator'] = RSIIndicator(df['PRC'], 14, False).rsi()
df['Stochastic_Oscillator'] = StochasticOscillator(df['ASKHI'], df['BIDLO'], df['PRC'], 14, 3, True).stoch()
# print("3")
n_fast = 12
n_slow = 26
df['MACD'] = df['Slow_Exp_Moving_Avg'] - df['Fast_Exp_Moving_Avg']
# Bollinger Bands
df['BollingerB_UP'] = df['Moving_Avg'] + df['Moving_Std_Deviation']*2
df['BollingerB_DOWN'] = df['Moving_Avg'] - df['Moving_Std_Deviation']*2
if stocks_industry.get(ticker) != None:
industry = stocks_industry.get(ticker).lower().replace(" ", "_")
industry_df = pd.read_csv(f'../database/industry_info/{industry}_returns.csv')
industry_df.set_index('date', inplace=True, drop=True)
df['avg_industry_return'] = industry_df['avg_return']
df.dropna(inplace = True)
X = df[df.columns[~df.columns.isin(['1_day_return','TICKER'])]]
y = df.loc[:, '1_day_return']
except:
return jsonify({})
xgb = XGBClassifier(random_state=0, seed = 312)
xgb.fit(X.iloc[:-1], y.iloc[:-1])
predict_for = pd.DataFrame(X.iloc[date]).T
print(predict_for)
# print(xgb.predict(X))
answer = xgb.predict_proba(predict_for)[0]
prediction = xgb.predict(predict_for)[0]
confidence = max(answer) * 100
feature_scores = xgb._Booster.get_score()
descriptors = {'OPENPRC': 'The open price',
'SHROUT': 'The number of shares outstanding',
'Moving_Avg': 'The average price of the stock over a rolling window of 20 days',
'Ultimate_Oscillator': 'Larry Williams’ (1976) signal, a momentum oscillator designed to capture momentum across three different timeframes.',
'VOL': 'The volume traded today',
'Stochastic_Oscillator': 'Developed in the late 1950s by George Lane. The stochastic oscillator presents the location of the closing price of a stock in relation to the high and low range of the price of a stock over a period of time, typically a 14-day period.',
'BIDLO': 'The lowest bid price today',
'ASKHI': 'The highest asking price today',
'Slow_Exp_Moving_Avg': 'This is a first-order infinite impulse response filter that applies weighting factors which decrease exponentially. Basically, a fancy moving average.',
'return': 'The return of the stock today, i.e. perentage change from the price yesterday to the price today',
'RSI_Indicator': 'Compares the magnitude of recent gains and losses over a specified time period to measure speed and change of price movements of a security. It is primarily used to attempt to identify overbought or oversold conditions in the trading of an asset.',
'BollingerB_DOWN': 'Developed by John Bollinger, Bollinger Bands® are volatility bands placed above and below a moving average. Volatility is based on the standard deviation, which changes as volatility increases and decreases',
'BollingerB_UP': 'Developed by John Bollinger, Bollinger Bands® are volatility bands placed above and below a moving average. Volatility is based on the standard deviation, which changes as volatility increases and decreases',
'Momentum_Factor': 'Simple Momentum of the stock',
'PRC': 'The closing price of the stock',
'MACD': 'A trend-following momentum indicator that shows the relationship between two moving averages of prices. The MACD is calculated by subtracting the 26-day exponential moving average (EMA) from the 12-day EMA',
'Fast_Exp_Moving_Avg': 'This is a first-order infinite impulse response filter that applies weighting factors which decrease exponentially. Basically, a fancy moving average.',
'avg_industry_return': 'The Average returns of the industry the stock is a part of.',
'Moving_Std_Deviation': 'The standard deviation over a period of 20-days'
}
links = {'OPENPRC': '',
'SHROUT': '',
'Moving_Avg': 'https://en.wikipedia.org/wiki/Moving_average',
'Ultimate_Oscillator': 'http://stockcharts.com/school/doku.php?id=chart_school:technical_indicators:ultimate_oscillator',
'VOL': '',
'Stochastic_Oscillator': 'https://school.stockcharts.com/doku.php?id=technical_indicators:stochastic_oscillator_fast_slow_and_full',
'BIDLO': '',
'ASKHI': '',
'Slow_Exp_Moving_Avg': 'https://en.wikipedia.org/wiki/Moving_average#Exponential_moving_average',
'return': '',
'RSI_Indicator': 'https://www.investopedia.com/terms/r/rsi.asp',
'BollingerB_DOWN': 'https://school.stockcharts.com/doku.php?id=technical_indicators:bollinger_bands',
'BollingerB_UP': 'https://school.stockcharts.com/doku.php?id=technical_indicators:bollinger_bands',
'Momentum_Factor': 'https://school.stockcharts.com/doku.php?id=technical_indicators:rate_of_change_roc_and_momentum',
'PRC': '',
'MACD': 'https://www.investopedia.com/terms/m/macd.asp',
'Fast_Exp_Moving_Avg': 'https://en.wikipedia.org/wiki/Moving_average#Exponential_moving_average',
'avg_industry_return': '',
'Moving_Std_Deviation': 'https://en.wikipedia.org/wiki/Moving_average'
}
classes = ['High', 'High', 'High', 'Medium', 'Medium', 'Medium', 'Medium', 'Medium', 'Low', 'Low', 'Low', 'Low', 'Low', 'Low', 'Almost None', 'Almost None', 'Almost None', 'Almost None', 'Almost None', 'Almost None']
scores = [[feature, weight]for feature, weight in feature_scores.items()]
scores.sort(key = lambda x: x[1], reverse = True)
for i, score in enumerate(scores):
feature = score[0]
score[1] = str(score[1])
score.append(classes[i])
score.append(descriptors[feature])
score.append(links[feature])
results = {
'prediction' : str(prediction),
'confidence' : str(confidence),
'feature_importance': feature_scores,
'descriptions' : scores
}
end = time.time()
print(end - start)
return jsonify(results)
def add_momentum_ta(
df: pd.DataFrame,
high: str,
low: str,
close: str,
volume: str,
fillna: bool = False,
colprefix: str = "",
vectorized: bool = False,
) -> pd.DataFrame:
"""Add trend 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.
volume (str): Name of 'volume' 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.
"""
# Relative Strength Index (RSI)
df[f"{colprefix}momentum_rsi"] = RSIIndicator(close=df[close],
window=14,
fillna=fillna).rsi()
# Stoch RSI (StochRSI)
indicator_srsi = StochRSIIndicator(close=df[close],
window=14,
smooth1=3,
smooth2=3,
fillna=fillna)
df[f"{colprefix}momentum_stoch_rsi"] = indicator_srsi.stochrsi()
df[f"{colprefix}momentum_stoch_rsi_k"] = indicator_srsi.stochrsi_k()
df[f"{colprefix}momentum_stoch_rsi_d"] = indicator_srsi.stochrsi_d()
# TSI Indicator
df[f"{colprefix}momentum_tsi"] = TSIIndicator(close=df[close],
window_slow=25,
window_fast=13,
fillna=fillna).tsi()
# Ultimate Oscillator
df[f"{colprefix}momentum_uo"] = UltimateOscillator(
high=df[high],
low=df[low],
close=df[close],
window1=7,
window2=14,
window3=28,
weight1=4.0,
weight2=2.0,
weight3=1.0,
fillna=fillna,
).ultimate_oscillator()
# Stoch Indicator
indicator_so = StochasticOscillator(
high=df[high],
low=df[low],
close=df[close],
window=14,
smooth_window=3,
fillna=fillna,
)
df[f"{colprefix}momentum_stoch"] = indicator_so.stoch()
df[f"{colprefix}momentum_stoch_signal"] = indicator_so.stoch_signal()
# Williams R Indicator
df[f"{colprefix}momentum_wr"] = WilliamsRIndicator(
high=df[high], low=df[low], close=df[close], lbp=14,
fillna=fillna).williams_r()
# Awesome Oscillator
df[f"{colprefix}momentum_ao"] = AwesomeOscillatorIndicator(
high=df[high], low=df[low], window1=5, window2=34,
fillna=fillna).awesome_oscillator()
# Rate Of Change
df[f"{colprefix}momentum_roc"] = ROCIndicator(close=df[close],
window=12,
fillna=fillna).roc()
# Percentage Price Oscillator
indicator_ppo = PercentagePriceOscillator(close=df[close],
window_slow=26,
window_fast=12,
window_sign=9,
fillna=fillna)
df[f"{colprefix}momentum_ppo"] = indicator_ppo.ppo()
df[f"{colprefix}momentum_ppo_signal"] = indicator_ppo.ppo_signal()
df[f"{colprefix}momentum_ppo_hist"] = indicator_ppo.ppo_hist()
# Percentage Volume Oscillator
indicator_pvo = PercentageVolumeOscillator(volume=df[volume],
window_slow=26,
window_fast=12,
window_sign=9,
fillna=fillna)
df[f"{colprefix}momentum_pvo"] = indicator_pvo.pvo()
df[f"{colprefix}momentum_pvo_signal"] = indicator_pvo.pvo_signal()
df[f"{colprefix}momentum_pvo_hist"] = indicator_pvo.pvo_hist()
if not vectorized:
# KAMA
df[f"{colprefix}momentum_kama"] = KAMAIndicator(close=df[close],
window=10,
pow1=2,
pow2=30,
fillna=fillna).kama()
return df
def add_momentum_ta(df: pd.DataFrame,
high: str,
low: str,
close: str,
volume: str,
fillna: bool = False,
colprefix: str = "") -> pd.DataFrame:
"""Add trend 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.
"""
# Relative Strength Index (RSI)
df[f'{colprefix}momentum_rsi'] = RSIIndicator(close=df[close],
n=14,
fillna=fillna).rsi()
# TSI Indicator
df[f'{colprefix}momentum_tsi'] = TSIIndicator(close=df[close],
r=25,
s=13,
fillna=fillna).tsi()
# Ultimate Oscillator
df[f'{colprefix}momentum_uo'] = UltimateOscillator(high=df[high],
low=df[low],
close=df[close],
s=7,
m=14,
len=28,
ws=4.0,
wm=2.0,
wl=1.0,
fillna=fillna).uo()
# Stoch Indicator
indicator = StochasticOscillator(high=df[high],
low=df[low],
close=df[close],
n=14,
d_n=3,
fillna=fillna)
df[f'{colprefix}momentum_stoch'] = indicator.stoch()
df[f'{colprefix}momentum_stoch_signal'] = indicator.stoch_signal()
# Williams R Indicator
df[f'{colprefix}momentum_wr'] = WilliamsRIndicator(high=df[high],
low=df[low],
close=df[close],
lbp=14,
fillna=fillna).wr()
# Awesome Oscillator
df[f'{colprefix}momentum_ao'] = AwesomeOscillatorIndicator(
high=df[high], low=df[low], s=5, len=34, fillna=fillna).ao()
# KAMA
df[f'{colprefix}momentum_kama'] = KAMAIndicator(close=df[close],
n=10,
pow1=2,
pow2=30,
fillna=fillna).kama()
# Rate Of Change
df[f'{colprefix}momentum_roc'] = ROCIndicator(close=df[close],
n=12,
fillna=fillna).roc()
return df
def setUp(self):
self._df = pd.read_csv(self._filename, sep=',')
self._indicator = UltimateOscillator(
high=self._df['High'], low=self._df['Low'], close=self._df['Close'],
s=7, m=14, len=28, ws=4.0, wm=2.0, wl=1.0, fillna=False)