Skip to content

Commit

Permalink
累计收益计算更新
Browse files Browse the repository at this point in the history
  • Loading branch information
@wukan1986
wukan1986 committed Mar 18, 2024
1 parent d63c4e7 commit 25aeb9b
Show file tree
Hide file tree
Showing 2 changed files with 77 additions and 29 deletions.
102 changes: 75 additions & 27 deletions alphainspect/portfolio.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -4,11 +4,12 @@
from loguru import logger
from matplotlib import pyplot as plt

from alphainspect import _QUANTILE_, _DATE_, _ASSET_
from alphainspect import _QUANTILE_, _DATE_, _ASSET_, _WEIGHT_
from alphainspect.utils import cumulative_returns, plot_heatmap


def calc_cum_return_by_quantile(df_pl: pl.DataFrame, fwd_ret_1: str, period: int = 5) -> pd.DataFrame:
"""分层计算收益。分成N层，层内等权"""
q_max = df_pl.select(pl.max(_QUANTILE_)).to_series(0)[0]
rr = df_pl.pivot(index=_DATE_, columns=_ASSET_, values=fwd_ret_1, aggregate_function='first', sort_columns=True).sort(_DATE_)
qq = df_pl.pivot(index=_DATE_, columns=_ASSET_, values=_QUANTILE_, aggregate_function='first', sort_columns=True).sort(_DATE_)
Expand All @@ -22,17 +23,19 @@ def calc_cum_return_by_quantile(df_pl: pl.DataFrame, fwd_ret_1: str, period: int
for i in range(int(q_max) + 1):
# 等权
b = qq == i
d = b / b.sum(axis=1).reshape(-1, 1)
d[(d == 0).all(axis=1), :] = np.nan
w = b / b.sum(axis=1).reshape(-1, 1)
w[(w == 0).all(axis=1), :] = np.nan
# 权重绝对值和为1
out[f'G{i}'] = cumulative_returns(rr, d, funds=period, freq=period)
out[f'G{i}'] = cumulative_returns(rr, w, funds=period, freq=period)
# !!!直接减是错误的，因为两资金是独立的，资金减少的一份由于资金不足对冲比例已经不再是1:1
# out['spread'] = out[f'G{q_max}'] - out[f'G0']
logger.info('累计收益计算完成,period={}\n{}', period, out.tail(1).to_string())
return out


def calc_cum_return_spread(df_pl: pl.DataFrame, fwd_ret_1: str, period: int = 5) -> pd.DataFrame:
"""分层计算收益。分成N层，层内等权。
取Top层和Bottom层。比较不同的计算方法多空收益的区别"""
q_max = df_pl.select(pl.max(_QUANTILE_)).to_series(0)[0]
rr = df_pl.pivot(index=_DATE_, columns=_ASSET_, values=fwd_ret_1, aggregate_function='first', sort_columns=True).sort(_DATE_).fill_nan(0)
qq = df_pl.pivot(index=_DATE_, columns=_ASSET_, values=_QUANTILE_, aggregate_function='first', sort_columns=True).sort(_DATE_).fill_nan(-1)
Expand All @@ -45,31 +48,52 @@ def calc_cum_return_spread(df_pl: pl.DataFrame, fwd_ret_1: str, period: int = 5)
np.seterr(divide='ignore', invalid='ignore')

# 等权
b0 = qq == 0
b9 = qq == q_max
b0 = b0 / b0.sum(axis=1).reshape(-1, 1)
b0 = np.where(b0 == b0, b0, 0)
b9 = b9 / b9.sum(axis=1).reshape(-1, 1)
b9 = np.where(b9 == b9, b9, 0)
bb = (b9 - b0) / 2 # 除2，权重绝对值和一定要调整为1，否则后面会计算错误
w0 = qq == 0
w9 = qq == q_max
w0 = w0 / w0.sum(axis=1).reshape(-1, 1)
w0 = np.where(w0 == w0, w0, 0)
w9 = w9 / w9.sum(axis=1).reshape(-1, 1)
w9 = np.where(w9 == w9, w9, 0)
ww = (w9 - w0) / 2 # 除2，权重绝对值和一定要调整为1，否则后面会计算错误

# 整行都为0，将其设成nan，后面计算时用于判断是否为0
bb[(bb == 0).all(axis=1), :] = np.nan
b0[(b0 == 0).all(axis=1), :] = np.nan
b9[(b9 == 0).all(axis=1), :] = np.nan
ww[(ww == 0).all(axis=1), :] = np.nan
w0[(w0 == 0).all(axis=1), :] = np.nan
w9[(w9 == 0).all(axis=1), :] = np.nan

# 曲线的翻转
out['1-G0 w=+1'] = 1 - cumulative_returns(rr, b0, funds=period, freq=period)
out['1-G0,w=+1'] = 1 - cumulative_returns(rr, w0, funds=period, freq=period)
# 权重的翻转。资金发生了变化。如果资金不共享，无法完全对冲
out['G0-1 w=-1'] = cumulative_returns(rr, -b0, funds=period, freq=period) - 1
out['G0-1,w=-1'] = cumulative_returns(rr, -w0, funds=period, freq=period) - 1

out[f'G{q_max} w=+1'] = cumulative_returns(rr, b9, funds=period, freq=period)
out[f'G{q_max},w=+1'] = cumulative_returns(rr, w9, funds=period, freq=period)
# 资金是共享的，每次调仓时需要将资金平分成两份
out[f'G{q_max}~G0 w=+.5/-.5'] = cumulative_returns(rr, bb, funds=period, freq=period, init_cash=1.0)
out[f'G{q_max}~G0,w=+.5/-.5'] = cumulative_returns(rr, ww, funds=period, freq=period, init_cash=1.0)
logger.info('多空收益计算完成,period={}\n{}', period, out.tail(1).to_string())
return out


def calc_cum_return_weights(df_pl: pl.DataFrame, fwd_ret_1: str, period: int = 1) -> pd.DataFrame:
"""指定权重计算收益。不再分层计算。资金也不分份"""
rr = df_pl.pivot(index=_DATE_, columns=_ASSET_, values=fwd_ret_1, aggregate_function='first', sort_columns=True).sort(_DATE_)
ww = df_pl.pivot(index=_DATE_, columns=_ASSET_, values=_WEIGHT_, aggregate_function='first', sort_columns=True).sort(_DATE_)

out = pd.DataFrame(index=rr[_DATE_], columns=rr.columns[1:])
rr = rr.select(pl.exclude(_DATE_)).to_numpy() # 日收益
ww = ww.select(pl.exclude(_DATE_)).to_numpy() # 权重
logger.info('权重收益准备数据,period={}', period)

np.seterr(divide='ignore', invalid='ignore')

rr = np.where(rr == rr, rr, 0.0)
# 累计收益分资产，资金不共享
# 由于是每天换仓，所以不存在空头计算不准的问题
out[:] = np.cumprod(rr * ww + 1, axis=0)

logger.info('权重收益计算完成,period={}\n{}', period, out.tail(1).to_string())
return out


def plot_quantile_portfolio(df_pd: pd.DataFrame, fwd_ret_1: str, period: int = 5,
*,
axvlines=None, ax=None) -> None:
Expand All @@ -80,9 +104,9 @@ def plot_quantile_portfolio(df_pd: pd.DataFrame, fwd_ret_1: str, period: int = 5
ax.axvline(x=v, c="b", ls="--", lw=1)


def plot_portfolio_heatmap(df_pd: pd.DataFrame,
*,
group='G9', ax=None) -> None:
def plot_portfolio_heatmap_monthly(df_pd: pd.DataFrame,
*,
group='G9', ax=None) -> None:
"""月度热力图。可用于IC, 收益率等"""
out = pd.DataFrame(index=df_pd.index)
out['year'] = out.index.year
Expand All @@ -94,22 +118,46 @@ def plot_portfolio_heatmap(df_pd: pd.DataFrame,
plot_heatmap(out['cum_ret'].unstack(), title=f"{group},Monthly Return", ax=ax)


def create_portfolio_sheet(df_pl: pl.DataFrame,
fwd_ret_1: str,
period=5,
*,
axvlines=()) -> None:
def create_portfolio1_sheet(df_pl: pl.DataFrame,
fwd_ret_1: str,
period=5,
*,
axvlines=()) -> None:
"""分层累计收益图"""
# 分层累计收益
df_cum_ret = calc_cum_return_by_quantile(df_pl, fwd_ret_1, period)

fig, axes = plt.subplots(2, 1, figsize=(12, 9))
plot_quantile_portfolio(df_cum_ret, fwd_ret_1, period, axvlines=axvlines, ax=axes[0])
groups = df_cum_ret.columns[[0, -1]]
for i, g in enumerate(groups):
ax = plt.subplot(223 + i)
plot_portfolio_heatmap(df_cum_ret, group=g, ax=ax)
# 月度收益
plot_portfolio_heatmap_monthly(df_cum_ret, group=g, ax=ax)
fig.tight_layout()

# 多空累计收益
df_spread = calc_cum_return_spread(df_pl, fwd_ret_1, period)
fig, axes = plt.subplots(1, 1, figsize=(12, 9))
plot_quantile_portfolio(df_spread, fwd_ret_1, period, axvlines=axvlines, ax=axes)
fig.tight_layout()


def create_portfolio2_sheet(df_pl: pl.DataFrame,
fwd_ret_1: str,
*,
axvlines=()) -> None:
"""分资产收益。权重由外部指定，资金是隔离"""
# 各资产收益，如果资产数量过多，图会比较卡顿
df_cum_ret = calc_cum_return_weights(df_pl, fwd_ret_1, 1)

fig, axes = plt.subplots(2, 1, figsize=(12, 9), squeeze=False)
axes = axes.flatten()
# 分资产收益
plot_quantile_portfolio(df_cum_ret, fwd_ret_1, 1, axvlines=axvlines, ax=axes[0])

# 资产平均收益，相当于等权
s = df_cum_ret.mean(axis=1)
s.name = 'portfolio'
plot_quantile_portfolio(s, fwd_ret_1, 1, axvlines=axvlines, ax=axes[1])
fig.tight_layout()
4 changes: 2 additions & 2 deletions examples/demo2.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -13,7 +13,7 @@
import polars as pl

from alphainspect.ic import create_ic1_sheet
from alphainspect.portfolio import create_portfolio_sheet
from alphainspect.portfolio import create_portfolio1_sheet
from alphainspect.returns import create_returns_sheet
from alphainspect.turnover import create_turnover_sheet
from alphainspect.utils import with_factor_quantile
Expand All @@ -38,7 +38,7 @@
create_returns_sheet(df_output, factor, forward_returns)
# %%
fwd_ret_1 = 'RETURN_OO_1' # 计算净值必需提供1日收益率
create_portfolio_sheet(df_output, fwd_ret_1, period=5, axvlines=axvlines)
create_portfolio1_sheet(df_output, fwd_ret_1, period=5, axvlines=axvlines)
create_turnover_sheet(df_output, factor, periods=(1, 5, 10, 20), axvlines=axvlines)

plt.show()
Expand Down

0 comments on commit 25aeb9b

Please sign in to comment.