调仓日可指定任意日期

config-svrobo5.yml

@@ -86,6 +86,7 @@ portfolios: # 投组模块
     dividend-date: 15 #配息日，每月15号
     dividend-adjust-day: [1,4,7,10]  #每年的首个季度调整配息
     warehouse-frequency: 1  #每隔1个月调一次仓
+    warehouse-transfer-date: 1  #调仓日
   solver: # 解算器相关
     tol: 1E-10 # 误差满足条件
     navs: # 净值要求

config-svrobo6.yml

@@ -51,7 +51,7 @@ py-jftech:
     max-workers: ${MAX_PROCESS:4}
 basic: # 基础信息模块
   sync:
-    start-date: 1990-01-01 # 同步数据开始日期
+    start-date: 2018-08-26 # 同步数据开始日期
   datum: # 资料模块
     change:
       date: ${DATUM_CHANGE_DATE}
@@ -85,11 +85,12 @@ portfolios: # 投组模块
     dividend-date: 15 #配息日，每月15号
     dividend-adjust-day: [1,4,7,10]  #每年的首个季度调整配息
     warehouse-frequency: 1  #每隔1个月调一次仓
+    warehouse-transfer-date: 1  #调仓日
     redeem-list: [ 'TEUSAAU LX Equity', 'LIGTRAA ID Equity', 'TEMFHAC LX Equity', 'LUSHUAA ID Equity' ] #从持仓中的低风险资产“直接”按序赎回
   solver: # 解算器相关
-    model: arc # 结算模型 ARC ,PRR,  ~ 标准解算器
+    model: prr # 结算模型 ARC ,PRR,  ~ 标准解算器
     arc: on  #是否开启ARC
-    brr: 0.01 #误差补偿值
+    brr: 0.02 #误差补偿值
     trr: 3
     tol: 1E-10 # 误差满足条件
     navs: # 净值要求
@@ -98,9 +99,9 @@ portfolios: # 投组模块
       max-nan: # 最大缺失净值条件
         asset: 8 # 单一资产最多缺少多少交易日数据，则踢出资产池
         day: 0.5 # 单一交易日最多缺少百分之多少净值，则删除该交易日
-      risk: []  # 资产风险等级要求，可分开写也可以合并写，e.g. risk：[ 2, 3 ] 则表示 所有投组资产风险等级都是 2 或 3
-    LARC: [0.5, 0.1, 0.1, 0.1]  #低阈值
-    UARC: [0.7, 0.25, 0.25, 0.25] #高阈值
+      risk: [ ]  # 资产风险等级要求，可分开写也可以合并写，e.g. risk：[ 2, 3 ] 则表示 所有投组资产风险等级都是 2 或 3
+    LARC: [ 0.30, 0.00, 0.00 ]  #低阈值
+    UARC: [ 0.70, 0.70, 0.70 ] #高阈值
     matrix-rtn-days: 20 # 计算回报率矩阵时，回报率滚动天数
     asset-count: [5,5] # 投组资产个数。e.g. count 或 [min, max] 分别表示 最大最小都为count 或 最小为min 最大为max，另外这里也可以类似上面给不同风险等级分别配置
     mpt: # mpt计算相关
@@ -234,12 +235,12 @@ robo-executor: # 执行器相关
   use: ${ROBO_EXECUTOR:backtest} # 执行哪个执行器，优先取系统环境变量ROBO_EXECUTOR的值，默认backtest
   sync-data: ${SYNC_DATA:off} # 是否开启同步资料数据
   backtest: # 回测执行器相关
-    start-date: 2022-02-16 # 回测起始日期
-    end-date: 2023-01-03 # 回测截止日期
+    start-date: 2018-11-26 # 回测起始日期
+    end-date: 2019-01-13 # 回测截止日期
     sealing-period: 10 #调仓封闭期
-    start-step: ${BACKTEST_START_STEP:3} # 回测从哪一步开始执行 1:计算资产池；2:计算最优投组：3:计算再平衡信号以及持仓投组
+    start-step: ${BACKTEST_START_STEP:1} # 回测从哪一步开始执行 1:计算资产池；2:计算最优投组：3:计算再平衡信号以及持仓投组
     end-step: ${BACKTEST_END_STEP:3} # 回测从哪一步执行完成后结束执行 1:计算资产池；2:计算最优投组：3:计算再平衡信号以及持仓投组
-    clean-up: on
+    clean-up: off
   real: # 实盘执行器
     export: ${EXPORT_ENABLE:off} # 是否开启报告
     start-date: 2023-05-08 # 实盘开始时间

portfolios/holder.py

@@ -266,7 +266,7 @@ class InvTrustPortfoliosHolder(DividendPortfoliosHolder):
             #  若调仓当日，有基金产生配息
             share_nav = {x: fund_nav * w / navs[x] for x, w in weight.items()}
             share_nodiv_nav = {x: nav * w / nav_cals[x] for x, w in weight.items()}
-            if self.is_first_workday(day):
+            if self.is_transfer_workday(day):
                 div_forecast = asset_nav * self.month_dividend
         else:
             fund_av = self.init_nav
@@ -311,10 +311,11 @@ class InvTrustPortfoliosHolder(DividendPortfoliosHolder):
             'asset_nav': asset_nav,
         })
 
-    def is_first_workday(self, day):
-        # 获取当月第一天的日期
-        first_day = date(day.year, day.month, 1)
-        first_work_day = first_day if is_workday(first_day) else next_workday(first_day)
+    def is_transfer_workday(self, day):
+        transfer_date = self._config['warehouse-transfer-date']
+        # 获取当月第n天的日期
+        transfer_date = date(day.year, day.month, transfer_date)
+        first_work_day = transfer_date if is_workday(transfer_date) else next_workday(transfer_date)
         return day.day == first_work_day.day
 
     def no_rebalance(self, day, risk: PortfoliosRisk, last_nav):
@@ -352,7 +353,7 @@ class InvTrustPortfoliosHolder(DividendPortfoliosHolder):
         weight_nodiv_nav = format_weight(weight_nodiv_nav)
         asset_nav = fund_av
         div_forecast = last_nav['div_forecast']
-        if self.is_first_workday(day):
+        if self.is_transfer_workday(day):
             div_forecast = asset_nav * self.month_dividend
         rhp.insert({
             'date': day,

rebalance/base_signal.py

@@ -6,7 +6,7 @@ from functools import reduce
 from typing import List
 
 import pandas as pd
-from py_jftech import component, autowired, get_config, prev_workday
+from py_jftech import component, autowired, get_config, prev_workday, workday_range
 from py_jftech import is_workday
 
 from api import PortfoliosBuilder
@@ -48,7 +48,13 @@ class BaseRebalanceSignal(RebalanceSignal, ABC):
         signal = rrs.get_last_one(day, risk, SignalType.NORMAL, effective=None)
         if signal:
             frequency = get_config('portfolios')['holder']['warehouse-frequency']
-            date = pd.to_datetime(day.replace(day=1)) + pd.DateOffset(months=frequency)
+            transfer_date = get_config('portfolios')['holder']['warehouse-transfer-date']
+            date = pd.to_datetime(signal['date'].replace(day=transfer_date))
+            # 说明发生了跨月份问题
+            if signal['date'].day > transfer_date:
+                if rrs.get_count(risk=PortfoliosRisk.FT3, effective=True) > 1:
+                    date = date + pd.DateOffset(months=1)
+            date = date + pd.DateOffset(months=frequency)
             date = date - timedelta(days=1)
             # 指定周期末的工作日
             date = date if is_workday(date) else prev_workday(date)

robo_executor.py

@@ -37,12 +37,13 @@ class BacktestExecutor(RoboExecutor):
 
     @staticmethod
     def get_last_business_day(start_date, end_date):
-        start_date = prev_workday(start_date)
+        transfer_date = get_config('portfolios')['holder']['warehouse-transfer-date']
         # 生成日期范围并转换为DataFrame
-        dates = pd.date_range(start_date, end_date, freq='M')
-        if dates[0] != start_date:
-            dates = dates.insert(0, start_date)
+        dates = pd.date_range(start_date, end_date, freq='MS', closed='right')
+        dates = [pd.to_datetime(f"{date.year}-{date.month}-{transfer_date}") for date in dates]
+        dates.insert(0, start_date)
         df = pd.DataFrame({'dates': dates})
+        df['dates'] = df['dates'].apply(lambda x: prev_workday(x))
         result = []
         for i in range(0, len(df), get_config('portfolios')['holder']['warehouse-frequency']):
             result.append(df.iloc[i]['dates'])