Skip to content

A
alpha-mind
Commit 8a4af8cd authored by wegamekinglc's avatar wegamekinglc
Browse files
Options

using more options as optimal

parent 1214d5ec
Hide whitespace changes
Inline Side-by-side
Showing with 90 additions and 38 deletions
alphamind/strategy/strategy.py
View file @ 8a4af8cd
......@@ -271,6 +271,7 @@ class Strategy(object):
if __name__ == '__main__':
import os
from matplotlib import pyplot as plt
from dask.distributed import Client
from PyFin.api import CSQuantiles
......@@ -290,82 +291,133 @@ if __name__ == '__main__':
mpl.rcParams['font.sans-serif'] = ['SimHei']
mpl.rcParams['axes.unicode_minus'] = False
start_date = '2010-05-01'
end_date = '2018-05-17'
# Back test parameter settings
start_date = '2016-01-01'
end_date = '2018-06-11'
freq = '10b'
neutralized_risk = None
industry_name = 'sw_adj'
industry_level = 1
turn_over_target = 0.4
batch = 1
horizon = map_freq(freq)
weights_bandwidth = 0.02
universe = Universe('zz800')
dask_client = Client('10.63.6.176:8786')
data_source = os.environ['DB_URI']
benchmark_code = 300
method = 'risk_neutral'
# Model settings
alpha_factors = {
'f1': CSQuantiles(LAST('ILLIQUIDITY') * LAST('NegMktValue'), groups='sw1_adj'),
'f2': CSQuantiles('con_pe', groups='sw1_adj')
'ep_q_cs': CSQuantiles(LAST('ep_q'), groups='sw1_adj'),
'roe_q_cs': CSQuantiles(LAST('roe_q'), groups='sw1_adj'),
'SGRO_cs': CSQuantiles(LAST('SGRO'), groups='sw1_adj'),
'GREV_cs': CSQuantiles(LAST('GREV'), groups='sw1_adj'),
'con_peg_rolling_cs': CSQuantiles(LAST('con_peg_rolling'), groups='sw1_adj'),
'con_pe_rolling_order_cs': CSQuantiles(LAST('con_pe_rolling_order'), groups='sw1_adj'),
'IVR_cs': CSQuantiles(LAST('IVR'), groups='sw1_adj'),
'ILLIQUIDITY_cs': CSQuantiles(LAST('ILLIQUIDITY') * LAST('NegMktValue'), groups='sw1_adj'),
'DividendPaidRatio_cs': CSQuantiles(LAST('DividendPaidRatio'), groups='sw1_adj'),
}
weights = {'f1': 1., 'f2': 0.}
# alpha_model = XGBTrainer(objective='reg:linear',
# booster='gbtree',
# n_estimators=300,
# eval_sample=0.25,
# features=alpha_factors)
weights = dict(ep_q_cs=1.,
roe_q_cs=1.,
SGRO_cs=0.0,
GREV_cs=0.0,
con_peg_rolling_cs=-0.25,
con_pe_rolling_order_cs=-0.25,
IVR_cs=0.5,
ILLIQUIDITY_cs=0.5,
DividendPaidRatio_cs=0.5)
alpha_model = ConstLinearModel(features=alpha_factors, weights=weights)
data_meta = DataMeta(freq=freq,
universe=universe,
batch=1,
neutralized_risk=neutralized_risk,
neutralized_risk=None,
pre_process=None,
post_process=None,
warm_start=1)
data_source=data_source)
industries = industry_list('sw_adj', 1)
# Constraintes settings
total_risk_names = ['total', 'benchmark'] + industries
industry_names = industry_list(industry_name, industry_level)
constraint_risk = ['SIZE', 'BETA']
total_risk_names = constraint_risk + ['benchmark', 'total']
all_styles = risk_styles + industry_names + macro_styles
b_type = []
l_val = []
u_val = []
previous_pos = pd.DataFrame()
rets = []
turn_overs = []
leverags = []
for name in total_risk_names:
if name == 'total':
if name == 'benchmark':
b_type.append(BoundaryType.RELATIVE)
l_val.append(0.8)
u_val.append(1.0)
elif name == 'total':
b_type.append(BoundaryType.ABSOLUTE)
l_val.append(.0)
u_val.append(.0)
elif name == 'benchmark':
elif name == 'EARNYILD':
b_type.append(BoundaryType.ABSOLUTE)
l_val.append(0.00)
u_val.append(0.60)
elif name == 'GROWTH':
b_type.append(BoundaryType.ABSOLUTE)
l_val.append(-0.05)
u_val.append(0.05)
elif name == 'MOMENTUM':
b_type.append(BoundaryType.ABSOLUTE)
l_val.append(0.20)
u_val.append(0.20)
elif name == 'SIZE':
b_type.append(BoundaryType.ABSOLUTE)
l_val.append(-0.05)
u_val.append(0.05)
elif name == 'LIQUIDTY':
b_type.append(BoundaryType.ABSOLUTE)
l_val.append(-0.40)
u_val.append(-0.0)
elif benchmark_code == 905 and name not in ["计算机", "医药生物", "国防军工", "信息服务", "机械设备"] and name in industry_names:
b_type.append(BoundaryType.RELATIVE)
l_val.append(0.8)
u_val.append(1.0)
elif benchmark_code == 300 and name in ["银行", "保险", "证券", "多元金融"]:
b_type.append(BoundaryType.RELATIVE)
l_val.append(0.70)
u_val.append(0.90)
elif name in ["计算机", "医药生物", "国防军工", "信息服务", "机械设备"]:
b_type.append(BoundaryType.ABSOLUTE)
l_val.append(0.0)
u_val.append(0.05)
else:
b_type.append(BoundaryType.MAXABSREL)
l_val.append((0.00, 0.0))
u_val.append((0.00, 0.0))
b_type.append(BoundaryType.ABSOLUTE)
l_val.append(-0.002)
u_val.append(0.002)
bounds = create_box_bounds(total_risk_names, b_type, l_val, u_val)
running_setting = RunningSetting(lbound=None,
ubound=None,
weights_bandwidth=0.01,
rebalance_method='risk_neutral',
# Running settings
running_setting = RunningSetting(weights_bandwidth=weights_bandwidth,
rebalance_method=method,
bounds=bounds,
target_vol=0.05,
turn_over_target=0.4)
turn_over_target=turn_over_target)
# Strategy
strategy = Strategy(alpha_model,
data_meta,
universe=universe,
start_date=start_date,
end_date=end_date,
freq=freq,
benchmark=906,
dask_client=dask_client)
strategy.prepare_backtest_data()
benchmark=benchmark_code)
ret_df, positions = strategy.run(running_setting)
ret_df.rename(columns={'excess_return': '超额收益', 'turn_over': '换手率'}, inplace=True)
ret_df[['超额收益', '换手率']].cumsum().plot(secondary_y='换手率')
plt.title("原始ILLIQUIDITY因子")
plt.show()
positions.to_csv('d:/positions.csv', encoding='gbk')
strategy.prepare_backtest_data()
ret_df, positions = strategy.run(running_setting=running_setting)
\ No newline at end of file
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment