# %%time
sampling_date = first_sampling_date
while sampling_date <= last_sampling_date:
# get samples
df_historic = data.load_historic(sampling_date=sampling_date,
column="retadj")
df_historic -= df_rf.loc[df_historic.index].values
# calculate stats
df_stats = pd.DataFrame(index=df_historic.columns)
df_stats["ret_excess"] = (1 + df_historic).prod() - 1
df_stats["var_annual"] = df_historic.var() * 252
if sampling_date < last_sampling_date:
# get excess return samples
df_future = data.load_future(sampling_date=sampling_date,
column="retadj")
df_future -= df_rf.loc[df_future.index].values
# slice expanding window
df_expanding_estimates = pd.DataFrame(index=df_future.columns)
for window_length in range(1, 13):
end_date = sampling_date + relativedelta(months=window_length,
day=31)
df_window = df_future[df_future.index <= end_date]
# calculate stats in window
df_stats["ret_excess_next{}M".format(
window_length)] = (1 + df_window).prod() - 1
df_stats["var_annual_next{}M".format(
window_length)] = df_window.var() * 252
data.store(
data=df_decomposition,
path="samples/{:%Y-%m-%d}/historic_daily.csv".format(sampling_date),
)
# increment monthly end of month
print("Completed decomposition at {:%Y-%m-%d}".format(sampling_date))
sampling_date += relativedelta(months=1, day=31)
# ### Forward part as expanding window
# %%time
sampling_date = first_sampling_date
while sampling_date < last_sampling_date:
# load betas
df_var = data.load_future(sampling_date=sampling_date, column="var")
spy_data = df_spy.loc[df_var.index]
# slice expanding window
df_expanding_decompositions = pd.DataFrame(index=df_var.unstack().index)
for window_length in range(1, 13):
end_date = sampling_date + relativedelta(months=window_length, day=31)
df_window = df_var[df_var.index <= end_date]
spy_window = spy_data[spy_data.index <= end_date]
betas_window = data.load_asset_estimates(
sampling_date=sampling_date,
columns=["spy_capm_spy_next{}M".format(window_length)],
)
# decompose
df_decomposition = decompose_variance(df_window, betas_window, spy_window)
)
# increment monthly end of month
print("Completed historic return factor model estimation at {:%Y-%m-%d}".
format(sampling_date))
sampling_date += relativedelta(months=1, day=31)
# %% [markdown]
# ### Forward part as expanding window
# %%
# %%time
sampling_date = first_sampling_date
while sampling_date < last_sampling_date:
# get excess return samples
df_future = data.load_future(sampling_date=sampling_date, column="retadj")
df_future -= df_rf.loc[df_future.index].values
# slice expanding window
df_expanding_estimates = pd.DataFrame(index=df_future.columns)
for window_length in range(1, 13):
if (months_difference(end_date=last_sampling_date,
start_date=sampling_date) >= window_length):
end_date = sampling_date + relativedelta(months=window_length,
day=31)
df_window = df_future[df_future.index <= end_date]
# estimate models in window
df_estimates, df_residuals = estimate_models(ret_models, df_window)
# collect