def compute_beta(benchmark, ric, bool_print=False):
capm = stream_classes.capm_manager(benchmark, ric)
capm.load_timeseries()
capm.compute()
if bool_print:
print('------')
print(capm)
beta = capm.beta
return beta
def create_dataframe_top_correlations(portfolio, benchmark, universe):
df = pd.DataFrame()
# columns = security + list of metrics
# number of rows = nb_rows
# your code comes here
columns = ['ric', 'correlation', 'abs_correlation', 'beta']
for i in universe:
if portfolio == i:
universe.remove(i)
index = list(range(len(universe)))
df = pd.DataFrame(index=index, columns=columns)
for i in universe:
n = universe.index(i)
df.iloc[n, df.columns.get_loc('ric')] = i
capm = stream_classes.capm_manager(portfolio, i)
b = stream_classes.capm_manager(benchmark, i)
capm.load_timeseries()
capm.compute()
corr = capm.correlation
b.load_timeseries()
b.compute()
beta = b.beta
df.iloc[n, df.columns.get_loc('correlation')] = corr
df.iloc[n, df.columns.get_loc('abs_correlation')] = abs(corr)
df.iloc[n, df.columns.get_loc('beta')] = beta
df.sort_values(by='abs_correlation',
ascending=False,
inplace=True,
ignore_index=True)
return df
import importlib import matplotlib.pyplot as plt from scipy.stats import skew, kurtosis, chi2, linregress # import our own files and reload import stream_functions importlib.reload(stream_functions) import stream_classes importlib.reload(stream_classes) # input parameters ric = 'EURUSD=X' # MT.AS SAN.MC BBVA.MC REP.MC VWS.CO EQNR.OL MXNUSD=X ^VIX GBPUSD=X CHFUSD=X benchmark = 'CHFUSD=X' # ^STOXX50E ^STOXX ^S&P500 ^NASDAQ ^FCHI ^GDAXI EURUSD=X capm = stream_classes.capm_manager(ric, benchmark) capm.load_timeseries() capm.compute() capm.scatterplot() capm.plot_normalised() capm.plot_dual_axes() print(capm) ''' References plots https://queirozf.com/entries/pandas-dataframe-plot-examples-with-matplotlib-pyplot '''
