def savefig(self):
returns = self.port.returns
model = 'Classic'
# 포트폴리오 수익률과 비중 저장
with open(
'returns-{}_{}.pkl'.format(self.risk_profile,
self.current_date), 'wb') as f:
pl.dump(returns, f)
with open('w-{}_{}.pkl'.format(self.risk_profile, self.current_date),
'wb') as f:
pl.dump(self.w, f)
# Efficient fronter 저장
points = 50 # Number of points of the frontier
frontier = self.port.efficient_frontier(model=model,
rm=self.rm,
points=points,
rf=self.rf,
hist=True)
label = 'Max-return Portfolio' if self.risk_profile == 4 else 'Min-risk Portfolio'
mu = self.port.mu # Expected returns
cov = self.port.cov # Covariance matrix
fig1, ax1 = plt.subplots()
ax1 = plf.plot_frontier(w_frontier=frontier,
mu=mu,
cov=cov,
returns=returns,
rm=self.rm,
rf=self.rf,
alpha=0.05,
cmap='viridis',
w=self.w,
label=label,
marker='*',
s=16,
c='r',
height=6,
width=10,
ax=ax1)
fig1.savefig('ef-{}_{}.png'.format(self.risk_profile,
self.current_date))
fig2, ax2 = plt.subplots(nrows=1, ncols=1)
# Plotting efficient frontier composition
ax2 = plf.plot_frontier_area(w_frontier=frontier,
cmap="tab20",
height=6,
width=10,
ax=None)
fig2.savefig('ef_area-{}_{}.png'.format(self.risk_profile,
self.current_date))
# Estimate optimal portfolio:
model = 'Classic' # Could be Classic (historical), BL (Black Litterman) or FM (Factor Model)
rm = 'MV' # Risk measure used, this time will be variance
obj = 'Sharpe' # Objective function, could be MinRisk, MaxRet, Utility or Sharpe
hist = True # Use historical scenarios for risk measures that depend on scenarios
rf = 0 # Risk free rate
l = 0 # Risk aversion factor, only useful when obj is 'Utility'
w = port.optimization(model=model, rm=rm, obj=obj, rf=rf, l=l, hist=hist)
print(w.T)
# Plotting the composition of the portfolio
ax = plf.plot_pie(w=w,
title='Sharpe Mean Variance',
others=0.05,
nrow=25,
cmap="tab20",
height=6,
width=10,
ax=None)
points = 50 # Number of points of the frontier
frontier = port.efficient_frontier(model=model,
rm=rm,
points=points,
rf=rf,
hist=hist)
print(frontier.T.head())
# Plotting the efficient frontier
label = 'Max Risk Adjusted Return Portfolio' # Title of point
mu = port.mu # Expected returns
cov = port.cov # Covariance matrix
def portfolio_optimizer(
rm='CVaR',
alpha=0.1,
risk_free=0.02,
strategy='sharpe_scale_patterns_day',
):
pd.options.display.float_format = '{:.1%}'.format
# 股票代码，我直接用我的选股程序获取选股列表。
position_signals = position(portfolio=strategy,
frequency='day',
market_type=QA.MARKET_TYPE.STOCK_CN,
verbose=False)
if (position_signals is not None) and \
(len(position_signals) > 0):
datestamp = position_signals.index[0][0]
position_signals_best = position_signals.loc[
position_signals[FLD.LEVERAGE_ONHOLD].gt(0.99), :]
if (len(position_signals_best) > 20):
position_signals = position_signals_best
else:
pass
codelist = position_signals.index.get_level_values(level=1).to_list()
# 获取股票中文名称，只是为了看得方便，交易策略并不需要股票中文名称
stock_names = QA.QA_fetch_stock_name(codelist)
codename = [stock_names.at[code, 'name'] for code in codelist]
codename_T = {codename[i]: codelist[i] for i in range(len(codelist))}
data_day = QA.QA_fetch_stock_day_adv(
codelist,
start='2014-01-01',
end='{}'.format(datetime.date.today())).to_qfq()
# 收益率序列
rets_jotion = data_day.add_func(kline_returns_func)
returns = pd.DataFrame(
columns=codelist,
index=sorted(
data_day.data.index.get_level_values(level=0).unique()))
for code in codelist:
returns[code] = rets_jotion.loc[(slice(None),
code), :].reset_index(level=[1],
drop=True)
returns = returns.fillna(0)
returns = returns.rename(
columns={codelist[i]: codename[i]
for i in range(len(codelist))})
import riskfolio.Portfolio as pf
# Building the portfolio object
port = pf.Portfolio(returns=returns)
# Calculating optimum portfolio
# Select method and estimate input parameters:
method_mu = 'hist' # Method to estimate expected returns based on historical data.
method_cov = 'hist' # Method to estimate covariance matrix based on historical data.
port.assets_stats(method_mu=method_mu, method_cov=method_cov, d=0.94)
## Estimate optimal portfolio:
model = 'Classic' # Could be Classic (historical), BL (Black Litterman) or FM (Factor Model)
obj = 'Sharpe' # Objective function, could be MinRisk, MaxRet, Utility or Sharpe
hist = True # Use historical scenarios for risk measures that depend on scenarios
rf = risk_free / 365 # Risk free rate
l = 0 # Risk aversion factor, only useful when obj is 'Utility'
port.alpha = alpha
# 暗色主题
plt.style.use('Solarize_Light2')
# 正常显示中文字体
plt.rcParams['font.sans-serif'] = ['Microsoft YaHei']
import riskfolio.PlotFunctions as plf
# Plotting the composition of the portfolio
w = port.optimization(model=model,
rm=rm,
obj=obj,
rf=rf,
l=l,
hist=hist)
opt_weights = w.copy()
opt_weights['code'] = opt_weights.apply(lambda x: codename_T[x.name],
axis=1)
opt_weights['name'] = opt_weights.apply(lambda x: x.name, axis=1)
opt_weights = opt_weights.set_index(['code'], drop=False)
print(u'交易日', datestamp)
show_verbose(opt_weights, obj, rm)
if (rm == 'CVaR'):
# Risk measure CVaR
title = 'Sharpe Mean CVaR'
elif (rm == 'MV'):
# Risk measure used, this time will be variance
title = 'Sharpe Mean Variance'
elif (rm == 'WR'):
title = 'Sharpe Mean WR'
elif (rm == 'Sortino'):
title = 'Sortino Mean WR'
else:
rm = 'CVaR'
title = 'Sharpe Mean CVaR'
ax = plf.plot_pie(w=w,
title=title,
others=0.05,
nrow=25,
cmap="tab20",
height=6,
width=10,
ax=None)
plt.show()
## Plotting efficient frontier composition
#points = 10 # Number of points of the frontier
#frontier = port.efficient_frontier(model=model, rm=rm, points=points,
#rf=rf, hist=hist)
##print(frontier.T.head())
#ax = plf.plot_frontier_area(w_frontier=frontier, cmap="tab20", height=6,
#width=10, ax=None)
#plt.show()
else:
print(u'没有可用的选股数据。')
#%%
label = "Max Risk Adjusted Return Portfolio"
mu = port.mu
cov = port.cov
returns = port.returns
ax = plf.plot_frontier(
w_frontier=w_5,
mu=mu,
cov=cov,
returns=returns,
rm=rm,
rf=0,
alpha=0.01,
cmap="viridis",
w=w2_1,
label="Portfolio",
marker="*",
s=16,
c="r",
height=6,
width=10,
ax=None,
)
ax = plf.plot_pie(w=w2_1,
title="Portafolio",
height=6,
width=10,
cmap="tab20",
ax=None)
def jupyter_report(
returns,
w,
rm="MV",
rf=0,
alpha=0.05,
others=0.05,
nrow=25,
height=6,
width=14,
t_factor=252,
):
r"""
Create a matplotlib report with useful information to analyze risk and
profitability of investment portfolios.
Parameters
----------
returns : DataFrame
Assets returns.
w : DataFrame of shape (n_assets, 1)
Portfolio weights.
rm : str, optional
Risk measure used to estimate risk contribution.
The default is 'MV'. Posible values are:
- 'MV': Standard Deviation.
- 'MAD': Mean Absolute Deviation.
- 'MSV': Semi Standard Deviation.
- 'FLPM': First Lower Partial Moment (Omega Ratio).
- 'SLPM': Second Lower Partial Moment (Sortino Ratio).
- 'CVaR': Conditional Value at Risk.
- 'EVaR': Conditional Value at Risk.
- 'WR': Worst Realization (Minimax)
- 'MDD': Maximum Drawdown of uncompounded returns (Calmar Ratio).
- 'ADD': Average Drawdown of uncompounded returns.
- 'DaR': Drawdown at Risk of uncompounded returns.
- 'CDaR': Conditional Drawdown at Risk of uncompounded returns.
- 'UCI': Ulcer Index of uncompounded returns.
rf : float, optional
Risk free rate or minimum aceptable return. The default is 0.
alpha : float, optional
Significante level of VaR, CVaR, EVaR, DaR and CDaR.
The default is 0.05.
others : float, optional
Percentage of others section. The default is 0.05.
nrow : int, optional
Number of rows of the legend. The default is 25.
height : float, optional
Average height of charts in the image in inches. The default is 6.
width : float, optional
Width of the image in inches. The default is 14.
t_factor : float, optional
Factor used to annualize expected return and expected risks for
risk measures based on returns (not drawdowns). The default is 252.
.. math::
\begin{align}
\text{Annualized Return} & = \text{Return} \, \times \, \text{t_factor} \\
\text{Annualized Risk} & = \text{Risk} \, \times \, \sqrt{\text{t_factor}}
\end{align}
ax : matplotlib axis of size (6,1), optional
If provided, plot on this axis. The default is None.
Raises
------
ValueError
When the value cannot be calculated.
Returns
-------
ax : matplotlib axis
Returns the Axes object with the plot for further tweaking.
Example
-------
::
ax = rp.jupyter_report(returns, w, rm='MV', rf=0, alpha=0.05, height=6, width=14,
others=0.05, nrow=25)
.. image:: images/Report_1.png
.. image:: images/Report_2.png
.. image:: images/Report_3.png
.. image:: images/Report_4.png
"""
cov = returns.cov()
nav = returns.cumsum()
fig, ax = plt.subplots(
nrows=6,
figsize=(width, height * 6),
gridspec_kw={"height_ratios": [2, 1, 1.5, 1, 1, 1]},
)
ax[0] = plf.plot_table(returns, w, MAR=rf, alpha=alpha, t_factor=t_factor, ax=ax[0])
ax[2] = plf.plot_pie(
w=w,
title="Portfolio Composition",
others=others,
nrow=nrow,
cmap="tab20",
ax=ax[2],
)
ax[3] = plf.plot_risk_con(
w=w, cov=cov, returns=returns, rm=rm, rf=rf, alpha=alpha, ax=ax[3]
)
ax[4] = plf.plot_hist(returns=returns, w=w, alpha=alpha, bins=50, ax=ax[4])
ax[[1, 5]] = plf.plot_drawdown(nav=nav, w=w, alpha=0.05, ax=ax[[1, 5]])
year = str(datetime.datetime.now().year)
title = "Riskfolio-Lib Report"
subtitle = "Copyright (c) 2020-" + year + ", Dany Cajas. All rights reserved."
fig.suptitle(title, fontsize="xx-large", y=1.011, fontweight="bold")
ax[0].set_title(subtitle, fontsize="large", ha="center", pad=10)
return ax