克隆策略

In [9]:

# 本代码由可视化策略环境自动生成 2021年6月24日09:05
# 本代码单元只能在可视化模式下编辑。您也可以拷贝代码，粘贴到新建的代码单元或者策略，然后修改。


# 回测引擎：初始化函数，只执行一次
def m19_initialize_bigquant_run(context):
    # 加载预测数据
    context.ranker_prediction = context.options['data'].read_df()
    start_date = context.ranker_prediction["date"].min()
    end_date = context.ranker_prediction["date"].max()
    # 系统已经设置了默认的交易手续费和滑点，要修改手续费可使用如下函数
    context.set_commission(PerOrder(buy_cost=0.0003, sell_cost=0.0013, min_cost=5))
    # 预测数据，通过options传入进来，使用 read_df 函数，加载到内存 (DataFrame)
    # 设置买入的股票数量，这里买入预测股票列表排名靠前的5只
    stock_count = 5
    # 每只的股票的权重，如下的权重分配会使得靠前的股票分配多一点的资金，[0.339160, 0.213986, 0.169580, ..]
    context.stock_weights = T.norm([1 / math.log(i + 2) for i in range(0, stock_count)])
    # 设置每只股票占用的最大资金比例
    context.max_cash_per_instrument = 0.2
    context.options['hold_days'] = 5
    
    # handle_data 函数是每根bar都会执行，最好不要执行 DataSource 的读取
    context.industry_df = DataSource("industry_CN_STOCK_A").read(start_date=start_date, end_date=end_date, fields=["industry_sw_level1"])
    

# 回测引擎：每日数据处理函数，每天执行一次
def m19_handle_data_bigquant_run(context, data):
    
    # 按日期过滤得到今日的预测数据
    today = data.current_dt.strftime('%Y-%m-%d')
    ranker_prediction = context.ranker_prediction[context.ranker_prediction.date == today]

    # 1. 资金分配
    # 平均持仓时间是hold_days，每日都将买入股票，每日预期使用 1/hold_days 的资金
    # 实际操作中，会存在一定的买入误差，所以在前hold_days天，等量使用资金；之后，尽量使用剩余资金（这里设置最多用等量的1.5倍）
    is_staging = context.trading_day_index < context.options['hold_days'] # 是否在建仓期间（前 hold_days 天）
    cash_avg = context.portfolio.portfolio_value / context.options['hold_days']
    cash_for_buy = min(context.portfolio.cash, (1 if is_staging else 1.5) * cash_avg)
    cash_for_sell = cash_avg - (context.portfolio.cash - cash_for_buy)
    positions = {e.symbol: p.amount * p.last_sale_price
                 for e, p in context.portfolio.positions.items()}

    # 2. 生成卖出订单：hold_days天之后才开始卖出；对持仓的股票，按机器学习算法预测的排序末位淘汰
    if not is_staging and cash_for_sell > 0:
        equities = {e.symbol: e for e, p in context.portfolio.positions.items()}
        instruments = list(reversed(list(ranker_prediction.instrument[ranker_prediction.instrument.apply(
                lambda x: x in equities)])))

        for instrument in instruments:
            context.order_target(context.symbol(instrument), 0)
            cash_for_sell -= positions[instrument]
            if cash_for_sell <= 0:
                break

    # 3. 生成买入订单：按机器学习算法预测的排序，买入前面的stock_count只股票
    buy_cash_weights = context.stock_weights
    buy_instruments = list(ranker_prediction.instrument[:len(buy_cash_weights)])
    max_cash_per_instrument = context.portfolio.portfolio_value * context.max_cash_per_instrument
    positions_df = pd.DataFrame({"instrument": positions.keys(), "percent": np.round(list(positions.values()) / context.portfolio.portfolio_value, 2)})
    industry_df = context.industry_df[context.industry_df["date"] == today]
    positions_df = positions_df.merge(industry_df, on=["instrument"], how="left")
    positions_df = positions_df.groupby("industry_sw_level1", as_index=False, group_keys=False).sum()
    print(f"{today}的行业持仓：>>>>>>>>>>>>>>>>>>>>>>>>")
    # 此处可能存在行业占比 21% 的情况，是由于账户总价值（资金+持仓市值）的变化
    print(positions_df)
    industries = list(positions_df["industry_sw_level1"])
    
    for i, instrument in enumerate(buy_instruments):
        cash = cash_for_buy * buy_cash_weights[i]
        if cash > max_cash_per_instrument - positions.get(instrument, 0):
            # 确保股票持仓量不会超过每次股票最大的占用资金量
            cash = max_cash_per_instrument - positions.get(instrument, 0)
        if cash > 0:
            # 判断如果买入股票，板块仓位是否超过 20%
            industry = list(industry_df[industry_df["instrument"] == instrument]["industry_sw_level1"])[0]
            if industry in industries:
                old_percent = positions_df[positions_df["industry_sw_level1"] == industry]["percent"].values[0]
                if old_percent >= 0.2:
                    print(f"{today}购买{instrument}，板块{industry}已有20%仓位，取消订单")
                    continue
                new_percent = old_percent + round(cash / context.portfolio.portfolio_value, 2)
                if new_percent > 0.2:
                    print(f"{today}购买{instrument}超出板块{industry}持仓20%，调整买入金额")
                    cash = context.portfolio.portfolio_value * (0.2 - old_percent)
            context.order_value(context.symbol(instrument), cash)
            if len(positions_df) > 0:
                positions_df.loc[positions_df["industry_sw_level1"] == industry, "percent"] += round(cash / context.portfolio.portfolio_value, 2)

# 回测引擎：准备数据，只执行一次
def m19_prepare_bigquant_run(context):
    pass


m1 = M.instruments.v2(
    start_date='2010-01-01',
    end_date='2015-01-01',
    market='CN_STOCK_A',
    instrument_list='',
    max_count=0
)

m2 = M.advanced_auto_labeler.v2(
    instruments=m1.data,
    label_expr="""# #号开始的表示注释
# 0. 每行一个，顺序执行，从第二个开始，可以使用label字段
# 1. 可用数据字段见 https://bigquant.com/docs/develop/datasource/deprecated/history_data.html
#   添加benchmark_前缀，可使用对应的benchmark数据
# 2. 可用操作符和函数见 `表达式引擎 <https://bigquant.com/docs/develop/bigexpr/usage.html>`_

# 计算收益：5日收盘价(作为卖出价格)除以明日开盘价(作为买入价格)
shift(close, -5) / shift(open, -1)

# 极值处理：用1%和99%分位的值做clip
clip(label, all_quantile(label, 0.01), all_quantile(label, 0.99))

# 将分数映射到分类，这里使用20个分类
all_wbins(label, 20)

# 过滤掉一字涨停的情况 (设置label为NaN，在后续处理和训练中会忽略NaN的label)
where(shift(high, -1) == shift(low, -1), NaN, label)
""",
    start_date='',
    end_date='',
    benchmark='000300.SHA',
    drop_na_label=True,
    cast_label_int=True
)

m3 = M.input_features.v1(
    features="""# #号开始的表示注释
# 多个特征，每行一个，可以包含基础特征和衍生特征
return_5
return_10
return_20
avg_amount_0/avg_amount_5
avg_amount_5/avg_amount_20
rank_avg_amount_0/rank_avg_amount_5
rank_avg_amount_5/rank_avg_amount_10
rank_return_0
rank_return_5
rank_return_10
rank_return_0/rank_return_5
rank_return_5/rank_return_10
pe_ttm_0
"""
)

m15 = M.general_feature_extractor.v7(
    instruments=m1.data,
    features=m3.data,
    start_date='',
    end_date='',
    before_start_days=90
)

m16 = M.derived_feature_extractor.v3(
    input_data=m15.data,
    features=m3.data,
    date_col='date',
    instrument_col='instrument',
    drop_na=False,
    remove_extra_columns=False
)

m7 = M.join.v3(
    data1=m2.data,
    data2=m16.data,
    on='date,instrument',
    how='inner',
    sort=False
)

m5 = M.dropnan.v2(
    input_data=m7.data
)

m4 = M.stock_ranker_train.v6(
    training_ds=m5.data,
    features=m3.data,
    learning_algorithm='排序',
    number_of_leaves=30,
    minimum_docs_per_leaf=1000,
    number_of_trees=20,
    learning_rate=0.1,
    max_bins=1023,
    feature_fraction=1,
    data_row_fraction=1,
    ndcg_discount_base=1,
    m_lazy_run=False
)

m9 = M.instruments.v2(
    start_date=T.live_run_param('trading_date', '2015-01-05'),
    end_date=T.live_run_param('trading_date', '2015-02-01'),
    market='CN_STOCK_A',
    instrument_list='',
    max_count=0
)

m17 = M.general_feature_extractor.v7(
    instruments=m9.data,
    features=m3.data,
    start_date='',
    end_date='',
    before_start_days=90
)

m18 = M.derived_feature_extractor.v3(
    input_data=m17.data,
    features=m3.data,
    date_col='date',
    instrument_col='instrument',
    drop_na=False,
    remove_extra_columns=False
)

m10 = M.dropnan.v2(
    input_data=m18.data
)

m8 = M.stock_ranker_predict.v5(
    model=m4.model,
    data=m10.data,
    m_lazy_run=False
)

m19 = M.trade.v4(
    instruments=m9.data,
    options_data=m8.predictions,
    start_date='',
    end_date='',
    initialize=m19_initialize_bigquant_run,
    handle_data=m19_handle_data_bigquant_run,
    prepare=m19_prepare_bigquant_run,
    volume_limit=0.025,
    order_price_field_buy='open',
    order_price_field_sell='close',
    capital_base=1000000,
    auto_cancel_non_tradable_orders=True,
    data_frequency='daily',
    price_type='真实价格',
    product_type='股票',
    plot_charts=True,
    backtest_only=False,
    benchmark='000300.SHA'
)

[2021-06-24 08:58:31.908853] INFO: moduleinvoker: instruments.v2 开始运行..

[2021-06-24 08:58:31.916401] INFO: moduleinvoker: 命中缓存

[2021-06-24 08:58:31.918488] INFO: moduleinvoker: instruments.v2 运行完成[0.009655s].

[2021-06-24 08:58:31.921310] INFO: moduleinvoker: advanced_auto_labeler.v2 开始运行..

[2021-06-24 08:58:31.926879] INFO: moduleinvoker: 命中缓存

[2021-06-24 08:58:31.929458] INFO: moduleinvoker: advanced_auto_labeler.v2 运行完成[0.008125s].

[2021-06-24 08:58:31.933391] INFO: moduleinvoker: input_features.v1 开始运行..

[2021-06-24 08:58:31.939701] INFO: moduleinvoker: 命中缓存

[2021-06-24 08:58:31.942225] INFO: moduleinvoker: input_features.v1 运行完成[0.008852s].

[2021-06-24 08:58:31.961295] INFO: moduleinvoker: general_feature_extractor.v7 开始运行..

[2021-06-24 08:58:31.968366] INFO: moduleinvoker: 命中缓存

[2021-06-24 08:58:31.971317] INFO: moduleinvoker: general_feature_extractor.v7 运行完成[0.010061s].

[2021-06-24 08:58:31.975614] INFO: moduleinvoker: derived_feature_extractor.v3 开始运行..

[2021-06-24 08:58:31.981570] INFO: moduleinvoker: 命中缓存

[2021-06-24 08:58:31.984298] INFO: moduleinvoker: derived_feature_extractor.v3 运行完成[0.008678s].

[2021-06-24 08:58:31.988826] INFO: moduleinvoker: join.v3 开始运行..

[2021-06-24 08:58:32.001466] INFO: moduleinvoker: 命中缓存

[2021-06-24 08:58:32.004070] INFO: moduleinvoker: join.v3 运行完成[0.015233s].

[2021-06-24 08:58:32.007443] INFO: moduleinvoker: dropnan.v2 开始运行..

[2021-06-24 08:58:32.013506] INFO: moduleinvoker: 命中缓存

[2021-06-24 08:58:32.015088] INFO: moduleinvoker: dropnan.v2 运行完成[0.007646s].

[2021-06-24 08:58:32.017847] INFO: moduleinvoker: stock_ranker_train.v6 开始运行..

[2021-06-24 08:58:32.024522] INFO: moduleinvoker: 命中缓存

[2021-06-24 08:58:32.607212] INFO: moduleinvoker: stock_ranker_train.v6 运行完成[0.589337s].

[2021-06-24 08:58:32.611548] INFO: moduleinvoker: instruments.v2 开始运行..

[2021-06-24 08:58:32.617719] INFO: moduleinvoker: 命中缓存

[2021-06-24 08:58:32.619243] INFO: moduleinvoker: instruments.v2 运行完成[0.007709s].

[2021-06-24 08:58:32.625477] INFO: moduleinvoker: general_feature_extractor.v7 开始运行..

[2021-06-24 08:58:32.630992] INFO: moduleinvoker: 命中缓存

[2021-06-24 08:58:32.633279] INFO: moduleinvoker: general_feature_extractor.v7 运行完成[0.007795s].

[2021-06-24 08:58:32.638828] INFO: moduleinvoker: derived_feature_extractor.v3 开始运行..

[2021-06-24 08:58:32.645865] INFO: moduleinvoker: 命中缓存

[2021-06-24 08:58:32.648646] INFO: moduleinvoker: derived_feature_extractor.v3 运行完成[0.009813s].

[2021-06-24 08:58:32.652895] INFO: moduleinvoker: dropnan.v2 开始运行..

[2021-06-24 08:58:32.660564] INFO: moduleinvoker: 命中缓存

[2021-06-24 08:58:32.662593] INFO: moduleinvoker: dropnan.v2 运行完成[0.009648s].

[2021-06-24 08:58:32.668476] INFO: moduleinvoker: stock_ranker_predict.v5 开始运行..

[2021-06-24 08:58:32.677495] INFO: moduleinvoker: 命中缓存

[2021-06-24 08:58:32.680392] INFO: moduleinvoker: stock_ranker_predict.v5 运行完成[0.011871s].

[2021-06-24 08:58:32.714764] INFO: moduleinvoker: backtest.v8 开始运行..

[2021-06-24 08:58:32.760980] INFO: backtest: biglearning backtest:V8.5.0

[2021-06-24 08:58:32.763296] INFO: backtest: product_type:stock by specified

[2021-06-24 08:58:33.105216] INFO: moduleinvoker: cached.v2 开始运行..

[2021-06-24 08:58:33.116444] INFO: moduleinvoker: 命中缓存

[2021-06-24 08:58:33.118423] INFO: moduleinvoker: cached.v2 运行完成[0.013239s].

[2021-06-24 08:58:34.351177] INFO: algo: TradingAlgorithm V1.8.3

[2021-06-24 08:58:35.376012] INFO: algo: trading transform...

[2021-06-24 08:58:36.949825] INFO: Performance: Simulated 20 trading days out of 20.

[2021-06-24 08:58:36.952678] INFO: Performance: first open: 2015-01-05 09:30:00+00:00

[2021-06-24 08:58:36.955020] INFO: Performance: last close: 2015-01-30 15:00:00+00:00

[2021-06-24 08:58:39.205783] INFO: moduleinvoker: backtest.v8 运行完成[6.490988s].

[2021-06-24 08:58:39.208082] INFO: moduleinvoker: trade.v4 运行完成[6.523969s].

设置评估测试数据集，查看训练曲线
[视频教程]StockRanker训练曲线

bigcharts-data-start/{"__type":"tabs","__id":"bigchart-508d2817571d4f4e972ba31f83f91dc1"}/bigcharts-data-end

2015-01-05的行业持仓：>>>>>>>>>>>>>>>>>>>>>>>>
Empty DataFrame
Columns: [industry_sw_level1, percent]
Index: []
2015-01-06的行业持仓：>>>>>>>>>>>>>>>>>>>>>>>>
   industry_sw_level1  percent
0              220000     0.03
1              640000     0.07
2              710000     0.06
3              730000     0.04
2015-01-07的行业持仓：>>>>>>>>>>>>>>>>>>>>>>>>
   industry_sw_level1  percent
0              220000     0.06
1              370000     0.04
2              640000     0.07
3              710000     0.18
4              730000     0.04
2015-01-07购买300380.SZA超出板块710000持仓20%，调整买入金额
2015-01-07购买300367.SZA超出板块710000持仓20%，调整买入金额
2015-01-08的行业持仓：>>>>>>>>>>>>>>>>>>>>>>>>
   industry_sw_level1  percent
0              220000     0.09
1              270000     0.03
2              370000     0.10
3              640000     0.07
4              710000     0.20
5              730000     0.04
2015-01-08购买300209.SZA超出板块710000持仓20%，调整买入金额
2015-01-08购买603019.SHA超出板块710000持仓20%，调整买入金额
2015-01-09的行业持仓：>>>>>>>>>>>>>>>>>>>>>>>>
   industry_sw_level1  percent
0              220000     0.09
1              270000     0.09
2              370000     0.11
3              460000     0.03
4              630000     0.03
5              640000     0.07
6              710000     0.21
7              730000     0.04
2015-01-12的行业持仓：>>>>>>>>>>>>>>>>>>>>>>>>
   industry_sw_level1  percent
0              220000     0.12
1              270000     0.12
2              350000     0.02
3              370000     0.11
4              430000     0.06
5              460000     0.03
6              630000     0.07
7              640000     0.06
8              710000     0.22
9              730000     0.05
2015-01-13的行业持仓：>>>>>>>>>>>>>>>>>>>>>>>>
   industry_sw_level1  percent
0              220000     0.14
1              270000     0.13
2              350000     0.02
3              370000     0.11
4              430000     0.09
5              460000     0.03
6              630000     0.11
7              640000     0.08
8              710000     0.03
9              730000     0.05
2015-01-14的行业持仓：>>>>>>>>>>>>>>>>>>>>>>>>
   industry_sw_level1  percent
0              220000     0.08
1              240000     0.07
2              270000     0.13
3              350000     0.02
4              430000     0.12
5              460000     0.08
6              610000     0.03
7              630000     0.11
8              640000     0.09
9              650000     0.04
2015-01-15的行业持仓：>>>>>>>>>>>>>>>>>>>>>>>>
   industry_sw_level1  percent
0              220000     0.03
1              240000     0.20
2              270000     0.12
3              350000     0.02
4              420000     0.07
5              430000     0.07
6              610000     0.03
7              630000     0.11
8              640000     0.06
9              650000     0.04
2015-01-15购买000693.SZA，板块240000已有20%仓位，取消订单
2015-01-16的行业持仓：>>>>>>>>>>>>>>>>>>>>>>>>
   industry_sw_level1  percent
0              240000     0.13
1              270000     0.03
2              370000     0.04
3              420000     0.07
4              430000     0.08
5              630000     0.11
6              640000     0.06
7              650000     0.04
8              710000     0.04
2015-01-16购买600490.SHA超出板块240000持仓20%，调整买入金额
2015-01-16购买000693.SZA，板块240000已有20%仓位，取消订单
2015-01-19的行业持仓：>>>>>>>>>>>>>>>>>>>>>>>>
   industry_sw_level1  percent
0              220000     0.05
1              240000     0.20
2              270000     0.07
3              370000     0.05
4              420000     0.07
5              430000     0.12
6              630000     0.09
7              710000     0.04
2015-01-19购买000693.SZA，板块240000已有20%仓位，取消订单
2015-01-19购买002057.SZA，板块240000已有20%仓位，取消订单
2015-01-20的行业持仓：>>>>>>>>>>>>>>>>>>>>>>>>
   industry_sw_level1  percent
0              220000     0.05
1              230000     0.04
2              240000     0.20
3              270000     0.07
4              420000     0.03
5              430000     0.12
6              490000     0.14
2015-01-20购买000783.SZA超出板块490000持仓20%，调整买入金额
2015-01-20购买000693.SZA，板块240000已有20%仓位，取消订单
2015-01-20购买600490.SHA，板块240000已有20%仓位，取消订单
2015-01-21的行业持仓：>>>>>>>>>>>>>>>>>>>>>>>>
   industry_sw_level1  percent
0              240000     0.20
1              270000     0.03
2              430000     0.16
3              490000     0.21
4              640000     0.05
2015-01-21购买000809.SZA超出板块430000持仓20%，调整买入金额
2015-01-21购买600338.SHA，板块240000已有20%仓位，取消订单
2015-01-22的行业持仓：>>>>>>>>>>>>>>>>>>>>>>>>
   industry_sw_level1  percent
0              110000     0.04
1              240000     0.21
2              270000     0.08
3              340000     0.06
4              430000     0.20
5              640000     0.05
2015-01-23的行业持仓：>>>>>>>>>>>>>>>>>>>>>>>>
   industry_sw_level1  percent
0              110000     0.04
1              240000     0.05
2              270000     0.08
3              340000     0.06
4              360000     0.10
5              370000     0.04
6              410000     0.04
7              430000     0.20
8              450000     0.05
9              640000     0.11
2015-01-23购买002208.SZA，板块430000已有20%仓位，取消订单
2015-01-23购买000736.SZA，板块430000已有20%仓位，取消订单
2015-01-23购买600565.SHA，板块430000已有20%仓位，取消订单
2015-01-26的行业持仓：>>>>>>>>>>>>>>>>>>>>>>>>
   industry_sw_level1  percent
0              110000     0.04
1              270000     0.08
2              340000     0.06
3              360000     0.15
4              370000     0.04
5              410000     0.04
6              430000     0.04
7              450000     0.08
8              640000     0.07
2015-01-26购买002240.SZA超出板块360000持仓20%，调整买入金额
2015-01-27的行业持仓：>>>>>>>>>>>>>>>>>>>>>>>>
    industry_sw_level1  percent
0               110000     0.10
1               270000     0.08
2               330000     0.04
3               340000     0.06
4               360000     0.21
5               370000     0.04
6               410000     0.04
7               430000     0.04
8               450000     0.03
9               460000     0.04
10              640000     0.05
2015-01-27购买002240.SZA，板块360000已有20%仓位，取消订单
2015-01-28的行业持仓：>>>>>>>>>>>>>>>>>>>>>>>>
    industry_sw_level1  percent
0               110000     0.10
1               270000     0.03
2               330000     0.10
3               340000     0.06
4               360000     0.05
5               370000     0.08
6               410000     0.07
7               430000     0.04
8               450000     0.03
9               460000     0.04
10              640000     0.05
11              730000     0.05
2015-01-29的行业持仓：>>>>>>>>>>>>>>>>>>>>>>>>
   industry_sw_level1  percent
0              110000     0.10
1              330000     0.04
2              360000     0.20
3              370000     0.12
4              410000     0.03
5              450000     0.03
6              460000     0.04
7              630000     0.06
8              640000     0.09
9              730000     0.05
2015-01-30的行业持仓：>>>>>>>>>>>>>>>>>>>>>>>>
    industry_sw_level1  percent
0               110000     0.12
1               330000     0.05
2               360000     0.15
3               370000     0.16
4               410000     0.03
5               450000     0.03
6               460000     0.04
7               510000     0.03
8               630000     0.04
9               640000     0.09
10              730000     0.05

收益率19.72%
年化收益率865.66%
基准收益率-2.81%
阿尔法9.37
贝塔0.25
夏普比率10.39
胜率0.88
盈亏比4.07
收益波动率21.86%
信息比率0.47
最大回撤2.1%

bigcharts-data-start/{"__type":"tabs","__id":"bigchart-acfc5838280b42808b94fae357e46da8"}/bigcharts-data-end