克隆策略

StockRanker多因子选股策略¶

In [31]:

# 本代码由可视化策略环境自动生成 2021年11月17日 13:04
# 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。


# Python 代码入口函数，input_1/2/3 对应三个输入端，data_1/2/3 对应三个输出端
def m10_run_bigquant_run(input_1, input_2, input_3):
    # 示例代码如下。在这里编写您的代码
    df = input_1.read_df()#读取上一个模块的数据
    ins= list(df.instrument.unique())
    start = df.date.min()
    end = df.date.max()
    df1=D.history_data(instruments=ins,start_date=start,end_date=end,fields=['close_0','high_0','low_0','st_status_0'])
    df_final=pd.merge(df,df1,on=['date''instrument'])
    df_final=df_final[df_final['st_status_0']==0]
    df_final=df_final[df_final['low_0']>df_final['high_1']]
    df_final=df_final[df_final['close_0']>df_final['open_0']+0.02]
    data_1=DataSource.write_df(df_final)
    return Outputs(data_1=data_1, data_2=None, data_3=None)

# 后处理函数，可选。输入是主函数的输出，可以在这里对数据做处理，或者返回更友好的outputs数据格式。此函数输出不会被缓存。
def m10_post_run_bigquant_run(outputs):
    return outputs

# 回测引擎：初始化函数，只执行一次
def m4_initialize_bigquant_run(context):
    # 加载预测数据
    context.ranker_prediction = context.options['data'].read_df()

    # 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 = 3
    # 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 = 0.5
    context.options['hold_days'] = 1

# 回测引擎：每日数据处理函数，每天执行一次
def m4_handle_data_bigquant_run(context, data):
    # 按日期过滤得到今日的预测数据
    ranker_prediction = context.ranker_prediction[
        context.ranker_prediction.date == data.current_dt.strftime('%Y-%m-%d')]

    # 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.perf_tracker.position_tracker.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.perf_tracker.position_tracker.positions.items()}
        instruments = list(reversed(list(ranker_prediction.instrument[ranker_prediction.instrument.apply(
                lambda x: x in equities and not context.has_unfinished_sell_order(equities[x]))])))
        # print('rank order for sell %s' % instruments)
        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
    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:
            context.order_value(context.symbol(instrument), cash)

# 回测引擎：准备数据，只执行一次
def m4_prepare_bigquant_run(context):
    pass


g = T.Graph({

    'm1': 'M.instruments.v2',
    'm1.start_date': '2010-01-01',
    'm1.end_date': '2021-11-01',
    'm1.market': 'CN_STOCK_A',
    'm1.instrument_list': '',
    'm1.max_count': 0,

    'm2': 'M.advanced_auto_labeler.v2',
    'm2.instruments': T.Graph.OutputPort('m1.data'),
    'm2.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/data_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/big_expr.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)
""",
    'm2.start_date': '',
    'm2.end_date': '',
    'm2.benchmark': '000300.SHA',
    'm2.drop_na_label': True,
    'm2.cast_label_int': True,

    'm9': 'M.instruments.v2',
    'm9.start_date': T.live_run_param('trading_date', '2020-01-01'),
    'm9.end_date': T.live_run_param('trading_date', '2021-07-01'),
    'm9.market': 'CN_STOCK_A',
    'm9.instrument_list': '',
    'm9.max_count': 0,

    'm12': 'M.input_features.v1',
    'm12.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

rank(((high_0+low_0+close_0+open_0)/4-max((high_0+low_0+close_0+open_0)/4,15)))**delta(close_0,5)
 (ts_rank(volume_0/mean(volume_0,20),20)*ts_rank((-1*delta(close_0,7)),8))
fs_net_cash_flow_0/fs_net_income_0
mean(mf_net_pct_l_0,90)
mean(mf_net_pct_l_0,60)
mean(mf_net_pct_l_0,30)
avg_amount_180
mean(mf_net_pct_l_0,120)
ean(mf_net_pct_main_0,90)
sh_holder_avg_pct_0""",

    'm3': 'M.input_features.v1',
    'm3.features_ds': T.Graph.OutputPort('m12.data'),
    'm3.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

rank(((high_0+low_0+close_0+open_0)/4-max((high_0+low_0+close_0+open_0)/4,15)))**delta(close_0,5)
(ts_rank(volume_0/mean(volume_0,20),20)*ts_rank((-1*delta(close_0,7)),8))
fs_net_cash_flow_0/fs_net_income_0
mean(mf_net_pct_l_0,90)
mean(mf_net_pct_l_0,60)
mean(mf_net_pct_l_0,30)
avg_amount_180
mean(mf_net_pct_l_0,120)
ean(mf_net_pct_main_0,90)
sh_holder_avg_pct_0""",

    'm15': 'M.general_feature_extractor.v7',
    'm15.instruments': T.Graph.OutputPort('m1.data'),
    'm15.features': T.Graph.OutputPort('m3.data'),
    'm15.start_date': '',
    'm15.end_date': '',
    'm15.before_start_days': 0,

    'm16': 'M.derived_feature_extractor.v3',
    'm16.input_data': T.Graph.OutputPort('m15.data'),
    'm16.features': T.Graph.OutputPort('m3.data'),
    'm16.date_col': 'date',
    'm16.instrument_col': 'instrument',
    'm16.drop_na': False,
    'm16.remove_extra_columns': False,

    'm11': 'M.aa.v5',
    'm11.input_data': T.Graph.OutputPort('m16.data'),
    'm11.day_number': 100,

    'm7': 'M.join.v3',
    'm7.data1': T.Graph.OutputPort('m2.data'),
    'm7.data2': T.Graph.OutputPort('m11.data'),
    'm7.on': 'date,instrument',
    'm7.how': 'inner',
    'm7.sort': False,

    'm13': 'M.dropnan.v1',
    'm13.input_data': T.Graph.OutputPort('m7.data'),

    'm10': 'M.cached.v3',
    'm10.input_1': T.Graph.OutputPort('m13.data'),
    'm10.run': m10_run_bigquant_run,
    'm10.post_run': m10_post_run_bigquant_run,
    'm10.input_ports': '',
    'm10.params': '{}',
    'm10.output_ports': '',

    'm17': 'M.general_feature_extractor.v7',
    'm17.instruments': T.Graph.OutputPort('m9.data'),
    'm17.features': T.Graph.OutputPort('m3.data'),
    'm17.start_date': '',
    'm17.end_date': '',
    'm17.before_start_days': 60,

    'm18': 'M.derived_feature_extractor.v3',
    'm18.input_data': T.Graph.OutputPort('m17.data'),
    'm18.features': T.Graph.OutputPort('m3.data'),
    'm18.date_col': 'date',
    'm18.instrument_col': 'instrument',
    'm18.drop_na': False,
    'm18.remove_extra_columns': False,

    'm19': 'M.aa.v5',
    'm19.input_data': T.Graph.OutputPort('m18.data'),
    'm19.day_number': 100,

    'm14': 'M.dropnan.v1',
    'm14.input_data': T.Graph.OutputPort('m19.data'),

    'm6': 'M.stock_ranker_train.v5',
    'm6.training_ds': T.Graph.OutputPort('m10.data_1'),
    'm6.features': T.Graph.OutputPort('m12.data'),
    'm6.learning_algorithm': '排序',
    'm6.number_of_leaves': 30,
    'm6.minimum_docs_per_leaf': 1000,
    'm6.number_of_trees': 20,
    'm6.learning_rate': 0.1,
    'm6.max_bins': 1023,
    'm6.feature_fraction': 1,
    'm6.m_lazy_run': False,

    'm8': 'M.stock_ranker_predict.v5',
    'm8.model': T.Graph.OutputPort('m6.model'),
    'm8.data': T.Graph.OutputPort('m14.data'),
    'm8.m_lazy_run': False,

    'm4': 'M.trade.v4',
    'm4.instruments': T.Graph.OutputPort('m9.data'),
    'm4.options_data': T.Graph.OutputPort('m8.predictions'),
    'm4.start_date': '2020-01-01',
    'm4.end_date': '2021-01-01',
    'm4.initialize': m4_initialize_bigquant_run,
    'm4.handle_data': m4_handle_data_bigquant_run,
    'm4.prepare': m4_prepare_bigquant_run,
    'm4.volume_limit': 0.025,
    'm4.order_price_field_buy': 'open',
    'm4.order_price_field_sell': 'close',
    'm4.capital_base': 1000000,
    'm4.auto_cancel_non_tradable_orders': True,
    'm4.data_frequency': 'daily',
    'm4.price_type': '后复权',
    'm4.product_type': '股票',
    'm4.plot_charts': True,
    'm4.backtest_only': False,
    'm4.benchmark': '',
})

# g.run({})


def m5_run_bigquant_run(
    bq_graph,
    inputs,
    trading_days_market='CN', # 使用那个市场的交易日历, TODO
    train_instruments_mid='m1', # 训练数据 证券代码列表 模块id
    test_instruments_mid='m9', # 测试数据 证券代码列表 模块id
    predict_mid='m8', # 预测 模块id
    trade_mid='m4', # 回测 模块id
    start_date='2015-01-01', # 数据开始日期
    end_date=T.live_run_param('trading_date', '2021-01-01'), # 数据结束日期
    train_update_days=250, # 更新周期，按交易日计算，每多少天更新一次
    train_update_days_for_live=None, #模拟实盘模式下的更新周期，按交易日计算，每多少天更新一次。如果需要在模拟实盘阶段使用不同的模型更新周期，可以设置这个参数
    train_data_min_days=250, # 最小数据天数，按交易日计算，所以第一个滚动的结束日期是 从开始日期到开始日期+最小数据天数
    train_data_max_days=250, # 最大数据天数，按交易日计算，0，表示没有限制，否则每一个滚动的开始日期=max(此滚动的结束日期-最大数据天数, 开始日期
    rolling_count_for_live=1, #实盘模式下滚动次数，模拟实盘模式下，取最后多少次滚动。一般在模拟实盘模式下，只用到最后一次滚动训练的模型，这里可以设置为1；如果你的滚动训练数据时间段很短，以至于期间可能没有训练数据，这里可以设置大一点。0表示没有限制
):
    def merge_datasources(input_1):
        df_list = [ds[0].read_df().set_index('date').loc[ds[1]:].reset_index() for ds in input_1]
        df = pd.concat(df_list)
        instrument_data = {
            'start_date': df['date'].min().strftime('%Y-%m-%d'),
            'end_date': df['date'].max().strftime('%Y-%m-%d'),
            'instruments': list(set(df['instrument'])),
        }
        return Outputs(data=DataSource.write_df(df), instrument_data=DataSource.write_pickle(instrument_data))

    def gen_rolling_dates(trading_days_market, start_date, end_date, train_update_days, train_update_days_for_live, train_data_min_days, train_data_max_days, rolling_count_for_live):
        # 是否实盘模式
        tdays = list(D.trading_days(market=trading_days_market, start_date=start_date, end_date=end_date)['date'])
        is_live_run = T.live_run_param('trading_date', None) is not None

        if is_live_run and train_update_days_for_live:
            train_update_days = train_update_days_for_live

        rollings = []
        train_end_date = train_data_min_days
        while train_end_date < len(tdays):
            if train_data_max_days is not None and train_data_max_days > 0:
                train_start_date = max(train_end_date - train_data_max_days, 0)
            else:
                train_start_date = 0
            rollings.append({
                'train_start_date': tdays[train_start_date].strftime('%Y-%m-%d'),
                'train_end_date': tdays[train_end_date - 1].strftime('%Y-%m-%d'),
                'test_start_date': tdays[train_end_date].strftime('%Y-%m-%d'),
                'test_end_date': tdays[min(train_end_date + train_update_days, len(tdays)) - 1].strftime('%Y-%m-%d'),
            })
            train_end_date += train_update_days

        if not rollings:
            raise Exception('没有滚动需要执行，请检查配置')

        if is_live_run and rolling_count_for_live:
            rollings = rollings[-rolling_count_for_live:]

        return rollings

    g = bq_graph

    rolling_dates = gen_rolling_dates(
        trading_days_market, start_date, end_date, train_update_days, train_update_days_for_live, train_data_min_days, train_data_max_days, rolling_count_for_live)

    # 训练和预测
    results = []
    for rolling in rolling_dates:
        parameters = {}
        # 先禁用回测
        parameters[trade_mid + '.__enabled__'] = False
        parameters[train_instruments_mid + '.start_date'] = rolling['train_start_date']
        parameters[train_instruments_mid + '.end_date'] = rolling['train_end_date']
        parameters[test_instruments_mid + '.start_date'] = rolling['test_start_date']
        parameters[test_instruments_mid + '.end_date'] = rolling['test_end_date']
        # print('------ rolling_train:', parameters)
        results.append(g.run(parameters))

    # 合并预测结果并回测
    mx = M.cached.v3(run=merge_datasources, input_1=[[result[predict_mid].predictions, result[test_instruments_mid].data.read_pickle()['start_date']] for result in results])
    parameters = {}
    parameters['*.__enabled__'] = False
    parameters[trade_mid + '.__enabled__'] = True
    parameters[trade_mid + '.instruments'] = mx.instrument_data
    parameters[trade_mid + '.options_data'] = mx.data

    trade = g.run(parameters)

    return {'rollings': results, 'trade': trade}


m5 = M.hyper_rolling_train.v1(
    run=m5_run_bigquant_run,
    run_now=True,
    bq_graph=g
)

[2021-11-17 13:03:49.414531] INFO: moduleinvoker: instruments.v2 开始运行..

[2021-11-17 13:03:49.435869] INFO: moduleinvoker: 命中缓存

[2021-11-17 13:03:49.437892] INFO: moduleinvoker: instruments.v2 运行完成[0.023368s].

[2021-11-17 13:03:49.444044] INFO: moduleinvoker: instruments.v2 开始运行..

[2021-11-17 13:03:49.455356] INFO: moduleinvoker: 命中缓存

[2021-11-17 13:03:49.457694] INFO: moduleinvoker: instruments.v2 运行完成[0.013659s].

[2021-11-17 13:03:49.462889] INFO: moduleinvoker: input_features.v1 开始运行..

[2021-11-17 13:03:49.480354] INFO: moduleinvoker: 命中缓存

[2021-11-17 13:03:49.482369] INFO: moduleinvoker: input_features.v1 运行完成[0.019487s].

[2021-11-17 13:03:49.493882] INFO: moduleinvoker: advanced_auto_labeler.v2 开始运行..

[2021-11-17 13:03:49.504231] INFO: moduleinvoker: 命中缓存

[2021-11-17 13:03:49.506119] INFO: moduleinvoker: advanced_auto_labeler.v2 运行完成[0.012239s].

[2021-11-17 13:03:49.510567] INFO: moduleinvoker: input_features.v1 开始运行..

[2021-11-17 13:03:49.532254] INFO: moduleinvoker: 命中缓存

[2021-11-17 13:03:49.534405] INFO: moduleinvoker: input_features.v1 运行完成[0.023842s].

[2021-11-17 13:03:49.549076] INFO: moduleinvoker: general_feature_extractor.v7 开始运行..

[2021-11-17 13:03:49.558762] INFO: moduleinvoker: 命中缓存

[2021-11-17 13:03:49.560621] INFO: moduleinvoker: general_feature_extractor.v7 运行完成[0.01158s].

[2021-11-17 13:03:49.574191] INFO: moduleinvoker: general_feature_extractor.v7 开始运行..

[2021-11-17 13:03:49.583882] INFO: moduleinvoker: 命中缓存

[2021-11-17 13:03:49.585318] INFO: moduleinvoker: general_feature_extractor.v7 运行完成[0.011143s].

[2021-11-17 13:03:49.592795] INFO: moduleinvoker: derived_feature_extractor.v3 开始运行..

[2021-11-17 13:03:49.603881] INFO: moduleinvoker: 命中缓存

[2021-11-17 13:03:49.605378] INFO: moduleinvoker: derived_feature_extractor.v3 运行完成[0.01258s].

[2021-11-17 13:03:49.613709] INFO: moduleinvoker: derived_feature_extractor.v3 开始运行..

[2021-11-17 13:03:49.621060] INFO: moduleinvoker: 命中缓存

[2021-11-17 13:03:49.622523] INFO: moduleinvoker: derived_feature_extractor.v3 运行完成[0.008811s].

[2021-11-17 13:03:49.642083] INFO: moduleinvoker: aa.v5 开始运行..

[2021-11-17 13:03:49.657107] INFO: moduleinvoker: 命中缓存

[2021-11-17 13:03:49.658569] INFO: moduleinvoker: aa.v5 运行完成[0.016501s].

[2021-11-17 13:03:49.669256] INFO: moduleinvoker: aa.v5 开始运行..

[2021-11-17 13:03:49.675935] INFO: moduleinvoker: 命中缓存

[2021-11-17 13:03:49.677486] INFO: moduleinvoker: aa.v5 运行完成[0.008221s].

[2021-11-17 13:03:49.685394] INFO: moduleinvoker: dropnan.v1 开始运行..

[2021-11-17 13:03:49.695748] INFO: moduleinvoker: 命中缓存

[2021-11-17 13:03:49.697257] INFO: moduleinvoker: dropnan.v1 运行完成[0.011861s].

[2021-11-17 13:03:49.705516] INFO: moduleinvoker: join.v3 开始运行..

[2021-11-17 13:03:49.716913] INFO: moduleinvoker: 命中缓存

[2021-11-17 13:03:49.718544] INFO: moduleinvoker: join.v3 运行完成[0.013024s].

[2021-11-17 13:03:49.728064] INFO: moduleinvoker: dropnan.v1 开始运行..

[2021-11-17 13:03:49.741274] INFO: moduleinvoker: 命中缓存

[2021-11-17 13:03:49.743585] INFO: moduleinvoker: dropnan.v1 运行完成[0.01553s].

[2021-11-17 13:03:49.756201] INFO: moduleinvoker: cached.v3 开始运行..

[2021-11-17 13:03:50.096676] WARNING: bigdatasource: cannot find filed [close_0] table in field_table_map!

[2021-11-17 13:03:50.100631] WARNING: bigdatasource: cannot find filed [high_0] table in field_table_map!

[2021-11-17 13:03:50.103406] WARNING: bigdatasource: cannot find filed [st_status_0] table in field_table_map!

[2021-11-17 13:03:50.106249] WARNING: bigdatasource: cannot find filed [low_0] table in field_table_map!

[2021-11-17 13:03:50.111769] ERROR: moduleinvoker: module name: cached, module version: v3, trackeback: KeyError: 'instrument'

[2021-11-17 13:03:50.147505] ERROR: moduleinvoker: module name: hyper_rolling_train, module version: v1, trackeback: KeyError: 'instrument'

---------------------------------------------------------------------------
KeyError                                  Traceback (most recent call last)
<ipython-input-31-40d7eb170148> in <module>
    344 
    345 
--> 346 m5 = M.hyper_rolling_train.v1(
    347     run=m5_run_bigquant_run,
    348     run_now=True,

<ipython-input-31-40d7eb170148> in m5_run_bigquant_run(bq_graph, inputs, trading_days_market, train_instruments_mid, test_instruments_mid, predict_mid, trade_mid, start_date, end_date, train_update_days, train_update_days_for_live, train_data_min_days, train_data_max_days, rolling_count_for_live)
    329         parameters[test_instruments_mid + '.end_date'] = rolling['test_end_date']
    330         # print('------ rolling_train:', parameters)
--> 331         results.append(g.run(parameters))
    332 
    333     # 合并预测结果并回测

<ipython-input-31-40d7eb170148> in m10_run_bigquant_run(input_1, input_2, input_3)
     10     start = df.date.min()
     11     end = df.date.max()
---> 12     df1=D.history_data(instruments=ins,start_date=start,end_date=end,fields=['close_0','high_0','low_0','st_status_0'])
     13     df_final=pd.merge(df,df1,on=['date''instrument'])
     14     df_final=df_final[df_final['st_status_0']==0]

KeyError: 'instrument'