def get_min(df, window=20, complete_data=True):
min = df.rolling(window).min()
if complete_data:
min = complete_missing_values(min)
return min
def get_sma(df, window=20, complete_data=True):
df_sma = df.rolling(window).mean()
if complete_data:
return complete_missing_values(df_sma)
return df_sma
def get_volatility(df, window=20, complete_data=True):
vol = df.rolling(window).std()
if complete_data:
vol = complete_missing_values(vol)
return vol
def get_max(df, window=20, complete_data=True):
max = df.rolling(window).max()
if complete_data:
max = complete_missing_values(max)
return max
def get_moving_beta(df, ticker_base, window=20, complete_data=True):
df_data = df.copy()
cov = df_data.rolling(window=window).cov()
cov = cov[ticker_base].unstack(level=1)
benchmark = cov[[ticker_base]].copy()
cov = cov.drop(ticker_base, 1)
if complete_data:
cov['moving beta'] = complete_missing_values(cov / benchmark.values)
else:
cov['moving beta'] = cov / benchmark.values
return cov[['moving beta']]