复制链接

克隆策略

StockRanker多因子选股策略¶

In [3]:

import pandas as pd
pd.DataFrame([m6.model_id]).to_pickle('/home/bigquant/work/userlib/model_stockranker.csv')

In [15]:

# ss=pd.read_csv('model_stockranker.csv',encoding='utf-8')
ss=pd.read_csv('model_stockranker.csv')
# ss=pd.read_csv('model_stockranker.csv',encoding='gbk')
ss

Out[15]:

	��
0	_new_Index��pandas.core.indexes.range��
1	RangeIndex��}�(�name�N�start�K
2	h}�(hNhK
3	h}�(hNhK
4	�blocks�]�}�(�values�h%�mgr_locs��builtins...

In [14]:

# ss=pd.read_csv(open('model_stockranker.csv'))
# ss
# import pickle
# fr=open('model_stockranker.csv','rb')
# pickle.load(fr)

---------------------------------------------------------------------------
BQInputRejected                           Traceback (most recent call last)
BQInputRejected: 编译错误，3: 抱歉，平台暂时不支持此模块：pickle

In [ ]:

In [2]:

# 本代码由可视化策略环境自动生成 2023年2月22日 15:12
# 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 m5_run_bigquant_run(input_1, input_2, input_3):
    # 示例代码如下。在这里编写您的代码
    data = pd.read_pickle('/home/bigquant/work/userlib/model_stockranker.csv')
    model_ds = DataSource.write_pickle(data.iloc[0].values[0])
    return Outputs(data_1=model_ds)

# 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 m5_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 = 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

# 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


m1 = M.instruments.v2(
    start_date='2014-12-01',
    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/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)
""",
    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=0
)

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
)

m13 = M.dropnan.v1(
    input_data=m7.data
)

m6 = M.stock_ranker_train.v5(
    training_ds=m13.data,
    features=m3.data,
    learning_algorithm='排序',
    number_of_leaves=3,
    minimum_docs_per_leaf=100,
    number_of_trees=5,
    learning_rate=0.3,
    max_bins=1023,
    feature_fraction=1,
    m_lazy_run=False
)

m9 = M.instruments.v2(
    start_date=T.live_run_param('trading_date', '2015-01-01'),
    end_date=T.live_run_param('trading_date', '2015-07-01'),
    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=60
)

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
)

m14 = M.dropnan.v1(
    input_data=m18.data
)

m5 = M.cached.v3(
    run=m5_run_bigquant_run,
    post_run=m5_post_run_bigquant_run,
    input_ports='',
    params='{}',
    output_ports=''
)

m8 = M.stock_ranker_predict.v5(
    model=m5.data_1,
    data=m14.data,
    m_lazy_run=False
)

m4 = M.trade.v4(
    instruments=m9.data,
    options_data=m8.predictions,
    start_date='',
    end_date='',
    initialize=m4_initialize_bigquant_run,
    handle_data=m4_handle_data_bigquant_run,
    prepare=m4_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='后复权',
    product_type='股票',
    plot_charts=True,
    backtest_only=False,
    benchmark=''
)

[2023-02-22 14:38:09.544562] INFO: moduleinvoker: instruments.v2 开始运行..

[2023-02-22 14:38:09.659600] INFO: moduleinvoker: instruments.v2 运行完成[0.115049s].

[2023-02-22 14:38:09.671605] INFO: moduleinvoker: advanced_auto_labeler.v2 开始运行..

[2023-02-22 14:38:10.072549] INFO: 自动标注(股票): 加载历史数据: 53383 行

[2023-02-22 14:38:10.075044] INFO: 自动标注(股票): 开始标注 ..

[2023-02-22 14:38:12.640527] INFO: moduleinvoker: advanced_auto_labeler.v2 运行完成[2.968919s].

[2023-02-22 14:38:12.648962] INFO: moduleinvoker: input_features.v1 开始运行..

[2023-02-22 14:38:12.658032] INFO: moduleinvoker: 命中缓存

[2023-02-22 14:38:12.660889] INFO: moduleinvoker: input_features.v1 运行完成[0.011926s].

[2023-02-22 14:38:12.681872] INFO: moduleinvoker: general_feature_extractor.v7 开始运行..

[2023-02-22 14:38:13.314525] INFO: 基础特征抽取: 年份 2014, 特征行数=53383

[2023-02-22 14:38:13.701427] INFO: 基础特征抽取: 年份 2015, 特征行数=0

[2023-02-22 14:38:13.749274] INFO: 基础特征抽取: 总行数: 53383

[2023-02-22 14:38:13.755422] INFO: moduleinvoker: general_feature_extractor.v7 运行完成[1.073565s].

[2023-02-22 14:38:13.766992] INFO: moduleinvoker: derived_feature_extractor.v3 开始运行..

[2023-02-22 14:38:13.986021] INFO: derived_feature_extractor: 提取完成 avg_amount_0/avg_amount_5, 0.003s

[2023-02-22 14:38:13.990812] INFO: derived_feature_extractor: 提取完成 avg_amount_5/avg_amount_20, 0.002s

[2023-02-22 14:38:13.995410] INFO: derived_feature_extractor: 提取完成 rank_avg_amount_0/rank_avg_amount_5, 0.002s

[2023-02-22 14:38:13.999665] INFO: derived_feature_extractor: 提取完成 rank_avg_amount_5/rank_avg_amount_10, 0.002s

[2023-02-22 14:38:14.004904] INFO: derived_feature_extractor: 提取完成 rank_return_0/rank_return_5, 0.003s

[2023-02-22 14:38:14.044754] INFO: derived_feature_extractor: 提取完成 rank_return_5/rank_return_10, 0.037s

[2023-02-22 14:38:14.224267] INFO: derived_feature_extractor: /y_2014, 53383

[2023-02-22 14:38:14.325664] INFO: moduleinvoker: derived_feature_extractor.v3 运行完成[0.558655s].

[2023-02-22 14:38:14.336973] INFO: moduleinvoker: join.v3 开始运行..

[2023-02-22 14:38:14.739974] INFO: join: /y_2014, 行数=41014/53383, 耗时=0.229523s

[2023-02-22 14:38:14.779067] INFO: join: 最终行数: 41014

[2023-02-22 14:38:14.788985] INFO: moduleinvoker: join.v3 运行完成[0.452s].

[2023-02-22 14:38:14.799951] INFO: moduleinvoker: dropnan.v1 开始运行..

[2023-02-22 14:38:14.995228] INFO: dropnan: /y_2014, 40884/41014

[2023-02-22 14:38:15.037818] INFO: dropnan: 行数: 40884/41014

[2023-02-22 14:38:15.044574] INFO: moduleinvoker: dropnan.v1 运行完成[0.244631s].

[2023-02-22 14:38:15.066441] INFO: moduleinvoker: stock_ranker_train.v5 开始运行..

[2023-02-22 14:38:15.302979] INFO: StockRanker: 特征预处理 ..

[2023-02-22 14:38:15.405825] INFO: StockRanker: prepare data: training ..

[2023-02-22 14:38:15.560990] INFO: StockRanker: sort ..

[2023-02-22 14:38:16.245515] INFO: StockRanker训练: 7c14a61a 准备训练: 40884 行数

[2023-02-22 14:38:16.525638] INFO: StockRanker训练: 正在训练 ..

[2023-02-22 14:39:37.071502] INFO: moduleinvoker: stock_ranker_train.v5 运行完成[82.005069s].

[2023-02-22 14:39:37.082354] INFO: moduleinvoker: instruments.v2 开始运行..

[2023-02-22 14:39:37.166422] INFO: moduleinvoker: instruments.v2 运行完成[0.083987s].

[2023-02-22 14:39:37.184569] INFO: moduleinvoker: general_feature_extractor.v7 开始运行..

[2023-02-22 14:39:37.846360] INFO: 基础特征抽取: 年份 2014, 特征行数=99861

[2023-02-22 14:39:39.101171] INFO: 基础特征抽取: 年份 2015, 特征行数=281127

[2023-02-22 14:39:39.177665] INFO: 基础特征抽取: 总行数: 380988

[2023-02-22 14:39:39.189825] INFO: moduleinvoker: general_feature_extractor.v7 运行完成[2.005279s].

[2023-02-22 14:39:39.197910] INFO: moduleinvoker: derived_feature_extractor.v3 开始运行..

[2023-02-22 14:39:40.199426] INFO: derived_feature_extractor: 提取完成 avg_amount_0/avg_amount_5, 0.003s

[2023-02-22 14:39:40.204475] INFO: derived_feature_extractor: 提取完成 avg_amount_5/avg_amount_20, 0.003s

[2023-02-22 14:39:40.208738] INFO: derived_feature_extractor: 提取完成 rank_avg_amount_0/rank_avg_amount_5, 0.002s

[2023-02-22 14:39:40.212910] INFO: derived_feature_extractor: 提取完成 rank_avg_amount_5/rank_avg_amount_10, 0.002s

[2023-02-22 14:39:40.245324] INFO: derived_feature_extractor: 提取完成 rank_return_0/rank_return_5, 0.003s

[2023-02-22 14:39:40.249489] INFO: derived_feature_extractor: 提取完成 rank_return_5/rank_return_10, 0.002s

[2023-02-22 14:39:40.609497] INFO: derived_feature_extractor: /y_2014, 99861

[2023-02-22 14:39:41.311894] INFO: derived_feature_extractor: /y_2015, 281127

[2023-02-22 14:39:41.717051] INFO: moduleinvoker: derived_feature_extractor.v3 运行完成[2.519112s].

[2023-02-22 14:39:41.729497] INFO: moduleinvoker: dropnan.v1 开始运行..

[2023-02-22 14:39:42.136153] INFO: dropnan: /y_2014, 99326/99861

[2023-02-22 14:39:43.001156] INFO: dropnan: /y_2015, 277492/281127

[2023-02-22 14:39:43.107190] INFO: dropnan: 行数: 376818/380988

[2023-02-22 14:39:43.120328] INFO: moduleinvoker: dropnan.v1 运行完成[1.390826s].

[2023-02-22 14:39:43.142997] INFO: moduleinvoker: cached.v3 开始运行..

[2023-02-22 14:39:43.153957] INFO: moduleinvoker: 命中缓存

[2023-02-22 14:39:43.156726] INFO: moduleinvoker: cached.v3 运行完成[0.013733s].

[2023-02-22 14:39:43.174491] INFO: moduleinvoker: stock_ranker_predict.v5 开始运行..

[2023-02-22 14:39:43.722301] INFO: StockRanker预测: /y_2014 ..

[2023-02-22 14:39:44.498238] INFO: StockRanker预测: /y_2015 ..

[2023-02-22 14:39:45.704112] INFO: moduleinvoker: stock_ranker_predict.v5 运行完成[2.529604s].

[2023-02-22 14:39:50.218791] INFO: moduleinvoker: backtest.v8 开始运行..

[2023-02-22 14:39:50.226519] INFO: backtest: biglearning backtest:V8.6.3

[2023-02-22 14:39:50.228034] INFO: backtest: product_type:stock by specified

[2023-02-22 14:39:50.324122] INFO: moduleinvoker: cached.v2 开始运行..

[2023-02-22 14:39:55.309126] INFO: backtest: 读取股票行情完成:1119996

[2023-02-22 14:39:56.326005] INFO: moduleinvoker: cached.v2 运行完成[6.001917s].

[2023-02-22 14:40:03.081026] INFO: backtest: algo history_data=DataSource(8b4b889a4d3146589adb444991e17674T)

[2023-02-22 14:40:03.083463] INFO: algo: TradingAlgorithm V1.8.9

[2023-02-22 14:40:04.265151] INFO: algo: trading transform...

[2023-02-22 14:40:08.649235] INFO: Performance: Simulated 120 trading days out of 120.

[2023-02-22 14:40:08.651897] INFO: Performance: first open: 2015-01-05 09:30:00+00:00

[2023-02-22 14:40:08.654252] INFO: Performance: last close: 2015-07-01 15:00:00+00:00

[2023-02-22 14:40:11.291288] INFO: moduleinvoker: backtest.v8 运行完成[21.072489s].

[2023-02-22 14:40:11.293842] INFO: moduleinvoker: trade.v4 运行完成[25.579783s].

设置测试数据集，查看训练迭代过程的NDCG

bigcharts-data-start/{"__type":"tabs","__id":"bigchart-4d85b802b8a54344b649d75a760cdbbd"}/bigcharts-data-end

收益率170.68%
年化收益率709.4%
基准收益率20.36%
阿尔法5.23
贝塔0.74
夏普比率5.29
胜率0.71
盈亏比1.53
收益波动率40.71%
信息比率0.34
最大回撤23.58%

bigcharts-data-start/{"__type":"tabs","__id":"bigchart-e39cc56e07d04c12a492b694ddc66118"}/bigcharts-data-end