feature_extractors/fractional_differentiation.py

from fracdiff.sklearn import FracdiffStat
import pandas as pd
import numpy as np

def feature_fractional_differentiation(df: pd.DataFrame, period: int, is_log_return: bool) -> pd.Series:
    frac_diff = FracdiffStat(window = period)
    input_series = df["close"].to_numpy().reshape(-1, 1)
    result = frac_diff.fit_transform(input_series)
    return pd.Series(result.squeeze(), index = df.index)
feat(FeatureExtraction): added fractionally differentiated returns to remove lagged returns (#95 ) 2021-12-28 22:50:09 +01:00			`from fracdiff.sklearn import FracdiffStat`
			`import pandas as pd`
			`import numpy as np`

			`def feature_fractional_differentiation(df: pd.DataFrame, period: int, is_log_return: bool) -> pd.Series:`
feat(Data): added daily_glassnode DataCollection (#99 ) 2022-01-03 13:57:36 +01:00			`frac_diff = FracdiffStat(window = period)`
feat(FeatureExtraction): added fractionally differentiated returns to remove lagged returns (#95 ) 2021-12-28 22:50:09 +01:00			`input_series = df["close"].to_numpy().reshape(-1, 1)`
feat(Data): added daily_glassnode DataCollection (#99 ) 2022-01-03 13:57:36 +01:00			`result = frac_diff.fit_transform(input_series)`
			`return pd.Series(result.squeeze(), index = df.index)`