blade
/
crypto_quant
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

optimize orb strategy

This commit is contained in:
blade 2025-09-03 18:26:27 +08:00
parent f79834647d
commit 644df3a42f
5 changed files with 181 additions and 85 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
.gitignore vendored
View File

@ -4,3 +4,4 @@
core/db/__pycache__/*.pyc
core/biz/__pycache__/*.pyc
core/statistics/__pycache__/*.pyc
/data

0
core/biz/market_data_from_csv.py Normal file
View File

BIN
core/trade/__pycache__/orb_trade.cpython-312.pyc
View File

Binary file not shown.

97
core/trade/orb_trade.py
View File

@ -38,6 +38,8 @@ class ORBStrategy:
direction=None,
by_sar=False,
symbol_bar_data=None,
price_range_mean_as_R=False,
by_big_k=False,
):
"""
初始化ORB策略参数
@ -63,6 +65,8 @@ class ORBStrategy:
:param is_us_stock: 是否是美股
:param direction: 方向，None=自动，Long=多头，Short=空头
:param by_sar: 是否根据SAR指标生成信号，True=是，False=否
:param price_range_mean_as_R: 是否将价格振幅均值作为$R，True=是，False=否
:param by_big_k: 是否根据K线实体部分，亦即abs(open-close)超过high-low的50%，True=是，False=否
"""
logger.info(
f"初始化ORB策略参数：股票代码={symbol}，K线周期={bar}，开始日期={start_date}，结束日期={end_date}，初始账户资金={initial_capital}，最大杠杆倍数={max_leverage}，单次交易风险比例={risk_per_trade}，每股交易佣金={commission_per_share}"
@ -106,6 +110,17 @@ class ORBStrategy:
if self.by_sar:
self.sar_desc = "考虑SAR"
self.symbol_bar_data = symbol_bar_data
self.price_range_mean_as_R = price_range_mean_as_R
if self.price_range_mean_as_R:
self.price_range_mean_as_R_desc = "R为振幅均值"
else:
self.price_range_mean_as_R_desc = "R为entry减stop"
self.by_big_k = by_big_k
if self.by_big_k:
self.by_big_k_desc = "K线实体过50%"
else:
self.by_big_k_desc = "无K线要求"
def run(self):
"""
@ -127,7 +142,9 @@ class ORBStrategy:
:param end_date: 结束日期（格式：YYYY-MM-DD）
:param interval: K线周期（默认5分钟）
"""
logger.info(f"开始获取{self.symbol}数据：{self.start_date}至{self.end_date}，间隔{self.bar}")
logger.info(
f"开始获取{self.symbol}数据：{self.start_date}至{self.end_date}，间隔{self.bar}"
)
if self.symbol_bar_data is None or len(self.symbol_bar_data) == 0:
data = self.db_market_data.query_market_data_by_symbol_bar(
self.symbol, self.bar, start=self.start_date, end=self.end_date
@ -153,7 +170,6 @@ class ORBStrategy:
self.start_date = data["Date"].min().strftime("%Y-%m-%d")
# 最大data["Date"]
self.end_date = data["Date"].max().strftime("%Y-%m-%d")
self.data = data[
[
"symbol",
@ -169,6 +185,7 @@ class ORBStrategy:
]
].copy()
self.data.rename(columns={date_time_field: "date_time"}, inplace=True)
self.calculate_price_range_mean()
self.symbol_bar_data = self.data.copy()
else:
self.data = self.symbol_bar_data.copy()
@ -189,21 +206,24 @@ class ORBStrategy:
pass
logger.info(f"收盘价均值：{self.close_mean}")
logger.info(f"初始资金调整为：{self.initial_capital}")
logger.info(f"成功获取{self.symbol}数据：{len(self.data)}根{self.bar}K线,开始日期={self.start_date}，结束日期={self.end_date}")
logger.info(
f"成功获取{self.symbol}数据：{len(self.data)}根{self.bar}K线,开始日期={self.start_date}，结束日期={self.end_date}"
)
def calculate_shares(self, account_value, entry_price, stop_price):
def calculate_shares(self, account_value, entry_price, stop_price, risk_assumed):
"""
根据ORB公式计算交易股数
:param account_value: 当前账户价值（美元）
:param entry_price: 交易entry价格（第二根5分钟K线开盘价）
:param stop_price: 止损价格（多头=第一根K线最低价，空头=第一根K线最高价）
:param risk_assumed: 风险金额（$R）,根据price_range_mean_as_R决定
:return: 整数股数（Shares）
"""
logger.info(
f"开始计算交易股数：账户价值={account_value}，entry价格={entry_price}，止损价格={stop_price}"
)
# 计算单交易风险金额（$R）
risk_per_trade_dollar = abs(entry_price - stop_price) # 风险金额取绝对值
risk_per_trade_dollar = risk_assumed # 风险金额取绝对值
if risk_per_trade_dollar <= 0:
return 0 # 无风险时不交易
@ -247,6 +267,12 @@ class ORBStrategy:
close1 = first_candle["Close"]
sar_signal = first_candle["sar_signal"]
if high1 == low1:
continue
if self.by_big_k:
if (abs(open1 - close1) / (high1 - low1)) < 0.5:
continue
# 第二根5分钟K线（entry信号）
second_candle = daily_data.iloc[1]
entry_price = second_candle["Open"] # entry价格=第二根K线开盘价
@ -303,6 +329,17 @@ class ORBStrategy:
f"生成信号完成：共{len(signals_df)}个交易日，其中多头{sum(signals_df['Signal']=='Long')}次，空头{sum(signals_df['Signal']=='Short')}次"
)
def calculate_price_range_mean(self):
"""
计算价格振幅均值，振幅为最高价与最低价之差
计算价格振幅标准差
要求用滑动窗口: window_size=100计算均值，每次计算都包含当前行
返回一个新列，列名为"PriceRangeMean"
"""
self.data["PriceRange"] = self.data["High"] - self.data["Low"]
self.data["PriceRangeMean"] = self.data["PriceRange"].rolling(window=100).mean()
self.data["PriceRangeStd"] = self.data["PriceRange"].rolling(window=100).std()
def backtest(self):
"""
回测ORB策略
@ -343,7 +380,18 @@ class ORBStrategy:
stop_price = signal_row["StopPrice"]
high1 = signal_row["High1"]
low1 = signal_row["Low1"]
risk_assumed = abs(entry_price - stop_price) # 计算$R
price_range = signal_row["PriceRange"]
price_range_mean = signal_row["PriceRangeMean"]
price_range_std = signal_row["PriceRangeStd"]
# 计算$R
if (
self.price_range_mean_as_R
and price_range_mean is not None
and price_range_mean > 0
):
risk_assumed = price_range_mean
else:
risk_assumed = abs(entry_price - stop_price)
profit_target = (
entry_price + (risk_assumed * self.profit_target_multiple)
if signal == "Long"
@ -351,7 +399,9 @@ class ORBStrategy:
)
# 计算交易股数
shares = self.calculate_shares(account_value, entry_price, stop_price)
shares = self.calculate_shares(
account_value, entry_price, stop_price, risk_assumed
)
if shares == 0:
# 股数为0→不交易
equity_history.append(account_value)
@ -425,6 +475,8 @@ class ORBStrategy:
"TradeID": trade_id,
"Direction": self.direction_desc,
"BySar": self.sar_desc,
"PriceRangeMeanAsR": self.price_range_mean_as_R_desc,
"ByBigK": self.by_big_k_desc,
"Symbol": self.symbol,
"Bar": self.bar,
"Date": date,
@ -467,8 +519,12 @@ class ORBStrategy:
else 0
)
# 计算盈亏比
profit_sum = self.trades_df[self.trades_df["ProfitLoss"] > 0]["ProfitLoss"].sum()
loss_sum = abs(self.trades_df[self.trades_df["ProfitLoss"] < 0]["ProfitLoss"].sum())
profit_sum = self.trades_df[self.trades_df["ProfitLoss"] > 0][
"ProfitLoss"
].sum()
loss_sum = abs(
self.trades_df[self.trades_df["ProfitLoss"] < 0]["ProfitLoss"].sum()
)
if loss_sum == 0:
profit_loss_ratio = float("inf")
else:
@ -504,7 +560,9 @@ class ORBStrategy:
logger.info(f"最大单笔盈利：${self.trades_df['ProfitLoss'].max():.2f}")
self.trades_summary["最大单笔盈利$"] = self.trades_df["ProfitLoss"].max()
logger.info(f"最大单笔亏损：${abs(self.trades_df['ProfitLoss'].min()):.2f}")
self.trades_summary["最大单笔亏损$"] = abs(self.trades_df["ProfitLoss"].min())
self.trades_summary["最大单笔亏损$"] = abs(
self.trades_df["ProfitLoss"].min()
)
else:
logger.info("没有交易")
self.trades_summary_df = pd.DataFrame([self.trades_summary])
@ -516,6 +574,8 @@ class ORBStrategy:
self.trades_summary = {}
self.trades_summary["方向"] = self.direction_desc
self.trades_summary["根据SAR"] = self.sar_desc
self.trades_summary["R算法"] = self.price_range_mean_as_R_desc
self.trades_summary["K线条件"] = self.by_big_k_desc
self.trades_summary["股票代码"] = self.symbol
self.trades_summary["K线周期"] = self.bar
self.trades_summary["开始日期"] = self.start_date
@ -532,7 +592,6 @@ class ORBStrategy:
self.trades_summary["最大单笔盈利$"] = 0
self.trades_summary["最大单笔亏损$"] = 0
def create_equity_curve(self):
"""
创建账户净值曲线
@ -600,7 +659,11 @@ class ORBStrategy:
marker="s",
markersize=4,
)
plt.title(f"ORB曲线 {symbol} {bar} {self.direction_desc} {self.sar_desc}", fontsize=14, fontweight="bold")
plt.title(
f"{symbol} {bar} {self.direction_desc} {self.sar_desc} {self.price_range_mean_as_R_desc} {self.by_big_k_desc}",
fontsize=14,
fontweight="bold",
)
plt.xlabel("时间", fontsize=12)
plt.ylabel("涨跌变化", fontsize=12)
plt.legend(fontsize=11)
@ -641,9 +704,7 @@ class ORBStrategy:
fontsize=10,
)
plt.tight_layout()
self.chart_save_path = (
f"{self.output_chart_folder}/{symbol}_{bar}_{self.direction_desc}_{self.sar_desc}_orb_strategy_equity_curve.png"
)
self.chart_save_path = f"{self.output_chart_folder}/{symbol}_{bar}_{self.direction_desc}_{self.sar_desc}_{self.price_range_mean_as_R_desc}_{self.by_big_k_desc}_orb.png"
plt.savefig(self.chart_save_path, dpi=150, bbox_inches="tight")
plt.close()
@ -653,16 +714,14 @@ class ORBStrategy:
"""
start_date = self.start_date.replace("-", "")
end_date = self.end_date.replace("-", "")
output_file_name = f"orb_{self.symbol}_{self.bar}_{start_date}_{end_date}_{self.direction_desc}_{self.sar_desc}.xlsx"
output_file_name = f"orb_{self.symbol}_{self.bar}_{start_date}_{end_date}_{self.direction_desc}_{self.sar_desc}_{self.price_range_mean_as_R_desc}_{self.by_big_k_desc}.xlsx"
output_file_path = os.path.join(self.output_excel_folder, output_file_name)
logger.info(f"导出{output_file_path}")
with pd.ExcelWriter(output_file_path) as writer:
self.trades_df.to_excel(writer, sheet_name="交易明细", index=False)
self.trades_summary_df.to_excel(writer, sheet_name="交易总结", index=False)
if os.path.exists(self.chart_save_path):
charts_dict = {
"账户净值曲线": self.chart_save_path
}
charts_dict = {"账户净值曲线": self.chart_save_path}
self.output_chart_to_excel(output_file_path, charts_dict)
def output_chart_to_excel(self, excel_file_path: str, charts_dict: dict):

60
orb_trade_main.py
View File

@ -16,6 +16,8 @@ def main():
bar = "5m"
direction_list = [None, "Long", "Short"]
by_sar_list = [False, True]
price_range_mean_as_R_list = [False, True]
by_big_k_list = [False, True]
start_date = "2024-01-01"
end_date = datetime.now().strftime("%Y-%m-%d")
profit_target_multiple = 10
@ -30,6 +32,8 @@ def main():
for is_us_stock in is_us_stock_list:
for direction in direction_list:
for by_sar in by_sar_list:
for price_range_mean_as_R in price_range_mean_as_R_list:
for by_big_k in by_big_k_list:
if is_us_stock:
symbols = US_STOCK_MONITOR_CONFIG.get("volume_monitor", {}).get(
"symbols", ["QQQ"]
@ -40,7 +44,7 @@ def main():
)
for symbol in symbols:
logger.info(
f"开始回测 {symbol}, 交易周期:{bar}, 开始日期:{start_date}, 结束日期:{end_date}, 是否是美股:{is_us_stock}, 交易方向:{direction}, 是否使用SAR:{by_sar}"
f"开始回测 {symbol}, 交易周期:{bar}, 开始日期:{start_date}, 结束日期:{end_date}, 是否是美股:{is_us_stock}, 交易方向:{direction}, 是否使用SAR:{by_sar}, 是否使用R为entry减stop:{price_range_mean_as_R}, 是否使用K线实体过50%:{by_big_k}"
)
symbol_bar_data = None
found_symbol_bar_data = False
@ -67,6 +71,8 @@ def main():
risk_per_trade=risk_per_trade,
commission_per_share=commission_per_share,
symbol_bar_data=symbol_bar_data,
price_range_mean_as_R=price_range_mean_as_R,
by_big_k=by_big_k,
)
symbol_bar_data, trades_df, trades_summary_df = orb_strategy.run()
if symbol_bar_data is None or len(symbol_bar_data) == 0:
@ -97,13 +103,19 @@ def main():
writer, sheet_name="最大总收益率记录", index=False
)
statitics_dict["max_total_return_record_df_grouped_count"].to_excel(
writer, sheet_name="最大总收益率记录_方向和根据SAR的组合", index=False
writer, sheet_name="最大总收益率_所有条件组合", index=False
)
statitics_dict["max_total_return_record_df_direction_count"].to_excel(
writer, sheet_name="最大总收益率记录_方向", index=False
writer, sheet_name="最大总收益率_方向", index=False
)
statitics_dict["max_total_return_record_df_sar_count"].to_excel(
writer, sheet_name="最大总收益率记录_根据SAR", index=False
writer, sheet_name="最大总收益率_根据SAR", index=False
)
statitics_dict["max_total_return_record_df_R_count"].to_excel(
writer, sheet_name="最大总收益率_R算法", index=False
)
statitics_dict["max_total_return_record_df_K_count"].to_excel(
writer, sheet_name="最大总收益率_K线条件", index=False
)
chart_path = r"./output/trade_sandbox/orb_strategy/chart/"
os.makedirs(chart_path, exist_ok=True)
@ -156,6 +168,8 @@ def statistics_summary(trades_summary_df: pd.DataFrame):
summary["股票代码"] = symbol
summary["方向"] = max_total_return_record["方向"]
summary["根据SAR"] = max_total_return_record["根据SAR"]
summary["R算法"] = max_total_return_record["R算法"]
summary["K线条件"] = max_total_return_record["K线条件"]
summary["总收益率%"] = max_total_return_record["总收益率%"]
summary["自然收益率%"] = max_total_return_record["自然收益率%"]
max_total_return_record_list.append(summary)
@ -178,14 +192,14 @@ def statistics_summary(trades_summary_df: pd.DataFrame):
# 其它（如Series、list、dict、ndarray等）转字符串
return str(v)
for key_col in ["方向", "根据SAR"]:
for key_col in ["方向", "根据SAR", "R算法", "K线条件"]:
if key_col in max_total_return_record_df.columns:
max_total_return_record_df[key_col] = max_total_return_record_df[
key_col
].apply(_to_hashable_scalar)
# 分组统计
max_total_return_record_df_grouped_count = (
max_total_return_record_df.groupby(["方向", "根据SAR"], dropna=False)
max_total_return_record_df.groupby(["方向", "根据SAR", "R算法", "K线条件"], dropna=False)
.size()
.reset_index(name="数量")
)
@ -215,15 +229,39 @@ def statistics_summary(trades_summary_df: pd.DataFrame):
by="数量", ascending=False, inplace=True
)
max_total_return_record_df_sar_count.reset_index(drop=True, inplace=True)
# 统计R算法的记录数目
max_total_return_record_df_R_count = (
max_total_return_record_df.groupby(["R算法"], dropna=False)
.size()
.reset_index(name="数量")
)
max_total_return_record_df_R_count.sort_values(
by="数量", ascending=False, inplace=True
)
max_total_return_record_df_R_count.reset_index(drop=True, inplace=True)
# 统计K线条件的记录数目
max_total_return_record_df_K_count = (
max_total_return_record_df.groupby(["K线条件"], dropna=False)
.size()
.reset_index(name="数量")
)
max_total_return_record_df_K_count.sort_values(
by="数量", ascending=False, inplace=True
)
max_total_return_record_df_K_count.reset_index(drop=True, inplace=True)
else:
# 构造空结果，保证下游写入Excel不报错
max_total_return_record_df_grouped_count = pd.DataFrame(
columns=["方向", "根据SAR", "数量"]
columns=["方向", "根据SAR", "R算法", "K线条件", "数量"]
)
max_total_return_record_df_direction_count = pd.DataFrame(
columns=["方向", "数量"]
)
max_total_return_record_df_sar_count = pd.DataFrame(columns=["根据SAR", "数量"])
max_total_return_record_df_R_count = pd.DataFrame(columns=["R算法", "数量"])
max_total_return_record_df_K_count = pd.DataFrame(columns=["K线条件", "数量"])
result = {
"statistics_summary_df": statistics_summary_df,
@ -231,6 +269,8 @@ def statistics_summary(trades_summary_df: pd.DataFrame):
"max_total_return_record_df_grouped_count": max_total_return_record_df_grouped_count,
"max_total_return_record_df_direction_count": max_total_return_record_df_direction_count,
"max_total_return_record_df_sar_count": max_total_return_record_df_sar_count,
"max_total_return_record_df_R_count": max_total_return_record_df_R_count,
"max_total_return_record_df_K_count": max_total_return_record_df_K_count,
}
return result
@ -335,9 +375,5 @@ def test():
if __name__ == "__main__":
# main()
chart_path = r"./output/trade_sandbox/orb_strategy/chart/"
excel_file_path = r"./output/trade_sandbox/orb_strategy/excel/summary/orb_strategy_summary_20250902174203.xlsx"
copy_chart_to_excel(chart_path, excel_file_path)
main()
# test()