复制链接

克隆策略

StockRanker多因子选股策略¶

In [3]:

# 本代码由可视化策略环境自动生成 2022年6月25日 03:08
# 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 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=0.0001, 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 = 1
    # 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 = [ 0.5 ]
    # 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):
    # 获取当前持仓
    positions = {e.symbol: p.amount * p.last_sale_price
                 for e, p in context.portfolio.positions.items()}
    
    today = data.current_dt.strftime('%Y-%m-%d')
    # 按日期过滤得到今日的预测数据
    ranker_prediction = context.ranker_prediction[
        context.ranker_prediction.date == today]
#     try:
#     #大盘风控模块，读取风控数据    
#         benckmark_risk=ranker_prediction['bm_0'].values[0]
#         if benckmark_risk > 0:
#             for instrument in positions.keys():
#                 context.order_target(context.symbol(instrument), 0)
#                 print(today,'大盘风控止损触发,全仓卖出')
#                 return
#     except:
#         print('--！')
        
    #当risk为1时，市场有风险，全部平仓，不再执行其它操作    
    # 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')]
    #cash_for_buy = min(context.portfolio.portfolio_value/2,context.portfolio.cash)
    #cash_for_buy = context.portfolio.portfolio_value
    #print(ranker_prediction)
    #cash_for_buy = context.portfolio.portfolio_value
    cash_for_buy = context.portfolio.cash
    buy_instruments = list(ranker_prediction.instrument)
    sell_instruments = [instrument.symbol for instrument in context.portfolio.positions.keys()]
    to_buy = set(buy_instruments[:1]) - set(sell_instruments) 
    to_sell = set(sell_instruments) -  set(buy_instruments[:1])
   
    
    for instrument in to_sell:
        context.order_target(context.symbol(instrument), 0)
    for instrument in to_buy:
        context.order_value(context.symbol(instrument), cash_for_buy)

def m4_prepare_bigquant_run(context):


     # 获取st状态和涨跌停状态
    
    context.status_df = D.features(instruments =context.instruments,start_date = context.start_date, end_date = context.end_date, 
                           fields=['st_status_0','price_limit_status_0','price_limit_status_1'])


m1 = M.instruments.v2(
    start_date='2013-02-01',
    end_date='2019-10-31',
    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
# 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)
# 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(high, -2) / shift(open, -1)-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))

# 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)
#where(label>0.5, NaN, label)
#where(label<-0.5, NaN, label)
""",
    start_date='',
    end_date='',
    benchmark='000300.SHA',
    drop_na_label=True,
    cast_label_int=True
)

m3 = M.input_features.v1(
    features="""#短周期因子
cond1=(market_cap_float_0>10000000000)&\
(volume_0>mean(volume_0, 5))&\
(amount_0>100000000)&\
(turn_0>0.08)&\
(list_days_0>20)&\
(mf_net_amount_xl_0>18000000)&\
(close_0/adjust_factor_0>mean(close_0/adjust_factor_0, 10))

#排序)
cond2=sum(price_limit_status_0==3,2)>1

#量价
cond6=open_0>close_1
cond7=ta_macd(close_0,'long')
cond8=ta_ma(close_0,10, derive='long')

#其他因子
close_0>ts_max(close_0,56)#53：当日收盘价破 56天最高价（创新高）
ta_sma_10_0/ta_sma_30_0#56:   10天的sma线/30天的sma线
swing_volatility_10_0/swing_volatility_60_0 #59:   10天的波动率/60天的波动率
rank_return_3  #61:   3天收益率的 排名
mf_net_amount_0>mf_net_amount_1  #62:  判断 当日的资金流入净额>昨日资金流入净额
ta_bbands_middleband_28_0 #69：布林带28天均线
#短周期因子
"""
)

m15 = M.general_feature_extractor.v7(
    instruments=m1.data,
    features=m3.data,
    start_date='',
    end_date='',
    before_start_days=60
)

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.filter.v3(
    input_data=m7.data,
    expr='cond1 and cond6 ',
    output_left_data=False
)

m20 = M.chinaa_stock_filter.v1(
    input_data=m5.data,
    index_constituent_cond=['全部'],
    board_cond=['上证主板', '深证主板'],
    industry_cond=['交通运输', '休闲服务', '传媒/信息服务', '公用事业', '农林牧渔', '化工', '医药生物', '商业贸易', '国防军工', '家用电器', '建筑材料/建筑建材', '建筑装饰', '房地产', '有色金属', '机械设备', '汽车/交运设备', '电子', '电气设备', '纺织服装', '综合', '计算机', '轻工制造', '通信', '采掘', '钢铁', '食品饮料'],
    st_cond=['正常'],
    delist_cond=['非退市'],
    output_left_data=False
)

m13 = M.dropnan.v1(
    input_data=m20.data
)

m11 = M.sort.v5(
    input_ds=m13.data,
    sort_by='cond2',
    group_by='date',
    keep_columns='--',
    ascending=True
)

m6 = M.stock_ranker_train.v6(
    training_ds=m11.sorted_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,
    plot_charts=True,
    ndcg_discount_base=1,
    m_lazy_run=False
)

m9 = M.instruments.v2(
    start_date=T.live_run_param('trading_date', '2021-01-01'),
    end_date=T.live_run_param('trading_date', '2022-06-24'),
    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
)

m10 = M.filter.v3(
    input_data=m18.data,
    expr='cond1 and cond6 ',
    output_left_data=False
)

m21 = M.chinaa_stock_filter.v1(
    input_data=m10.data,
    index_constituent_cond=['全部'],
    board_cond=['上证主板', '深证主板'],
    industry_cond=['交通运输', '休闲服务', '传媒/信息服务', '公用事业', '农林牧渔', '化工', '医药生物', '商业贸易', '国防军工', '家用电器', '建筑材料/建筑建材', '建筑装饰', '房地产', '有色金属', '机械设备', '汽车/交运设备', '电子', '电气设备', '纺织服装', '综合', '计算机', '轻工制造', '通信', '采掘', '钢铁', '食品饮料'],
    st_cond=['正常'],
    delist_cond=['非退市'],
    output_left_data=False
)

m14 = M.dropnan.v1(
    input_data=m21.data
)

m12 = M.sort.v5(
    input_ds=m14.data,
    sort_by='cond2',
    group_by='date',
    keep_columns='--',
    ascending=True
)

m8 = M.stock_ranker_predict.v5(
    model=m6.model,
    data=m12.sorted_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,
    order_price_field_buy='open',
    order_price_field_sell='close',
    capital_base=100000,
    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.HIX'
)

[2022-06-25 03:06:12.530069] INFO: moduleinvoker: instruments.v2 开始运行..

[2022-06-25 03:06:12.539405] INFO: moduleinvoker: 命中缓存

[2022-06-25 03:06:12.541818] INFO: moduleinvoker: instruments.v2 运行完成[0.011752s].

[2022-06-25 03:06:12.550832] INFO: moduleinvoker: advanced_auto_labeler.v2 开始运行..

[2022-06-25 03:06:12.558596] INFO: moduleinvoker: 命中缓存

[2022-06-25 03:06:12.560546] INFO: moduleinvoker: advanced_auto_labeler.v2 运行完成[0.009729s].

[2022-06-25 03:06:12.567177] INFO: moduleinvoker: input_features.v1 开始运行..

[2022-06-25 03:06:12.574782] INFO: moduleinvoker: 命中缓存

[2022-06-25 03:06:12.576708] INFO: moduleinvoker: input_features.v1 运行完成[0.009556s].

[2022-06-25 03:06:12.593570] INFO: moduleinvoker: general_feature_extractor.v7 开始运行..

[2022-06-25 03:06:12.605572] INFO: moduleinvoker: 命中缓存

[2022-06-25 03:06:12.607492] INFO: moduleinvoker: general_feature_extractor.v7 运行完成[0.013949s].

[2022-06-25 03:06:12.615057] INFO: moduleinvoker: derived_feature_extractor.v3 开始运行..

[2022-06-25 03:06:12.622212] INFO: moduleinvoker: 命中缓存

[2022-06-25 03:06:12.624203] INFO: moduleinvoker: derived_feature_extractor.v3 运行完成[0.009138s].

[2022-06-25 03:06:12.631935] INFO: moduleinvoker: join.v3 开始运行..

[2022-06-25 03:06:22.360839] INFO: join: /y_2012, 行数=0/50379, 耗时=1.615519s

[2022-06-25 03:06:25.750590] INFO: join: /y_2013, 行数=515395/564168, 耗时=3.386273s

[2022-06-25 03:06:29.302522] INFO: join: /y_2014, 行数=567883/569948, 耗时=3.545726s

[2022-06-25 03:06:33.040672] INFO: join: /y_2015, 行数=560441/569698, 耗时=3.732319s

[2022-06-25 03:06:36.782299] INFO: join: /y_2016, 行数=637482/641546, 耗时=3.736372s

[2022-06-25 03:06:40.872160] INFO: join: /y_2017, 行数=738271/743233, 耗时=4.084208s

[2022-06-25 03:06:45.746605] INFO: join: /y_2018, 行数=813530/816987, 耗时=4.868501s

[2022-06-25 03:06:50.515646] INFO: join: /y_2019, 行数=714641/725092, 耗时=4.76118s

[2022-06-25 03:06:50.703046] INFO: join: 最终行数: 4547643

[2022-06-25 03:06:50.727788] INFO: moduleinvoker: join.v3 运行完成[38.095833s].

[2022-06-25 03:06:50.738739] INFO: moduleinvoker: filter.v3 开始运行..

[2022-06-25 03:06:50.756567] INFO: filter: 使用表达式 cond1 and cond6  过滤

[2022-06-25 03:06:50.891822] INFO: filter: 过滤 /y_2012, 0/0/0

[2022-06-25 03:06:51.445426] INFO: filter: 过滤 /y_2013, 2064/0/515395

[2022-06-25 03:06:52.077913] INFO: filter: 过滤 /y_2014, 3701/0/567883

[2022-06-25 03:06:52.677840] INFO: filter: 过滤 /y_2015, 12081/0/560441

[2022-06-25 03:06:53.392063] INFO: filter: 过滤 /y_2016, 5997/0/637482

[2022-06-25 03:06:54.149945] INFO: filter: 过滤 /y_2017, 4753/0/738271

[2022-06-25 03:06:54.991620] INFO: filter: 过滤 /y_2018, 3562/0/813530

[2022-06-25 03:06:55.768384] INFO: filter: 过滤 /y_2019, 4677/0/714641

[2022-06-25 03:06:55.798809] INFO: moduleinvoker: filter.v3 运行完成[5.060072s].

[2022-06-25 03:06:55.807327] INFO: moduleinvoker: chinaa_stock_filter.v1 开始运行..

[2022-06-25 03:06:56.815539] INFO: A股股票过滤: 过滤 /y_2013, 1561/0/2064

[2022-06-25 03:06:57.878505] INFO: A股股票过滤: 过滤 /y_2014, 2873/0/3701

[2022-06-25 03:06:59.115364] INFO: A股股票过滤: 过滤 /y_2015, 10010/0/12081

[2022-06-25 03:07:00.339004] INFO: A股股票过滤: 过滤 /y_2016, 4815/0/5997

[2022-06-25 03:07:01.796162] INFO: A股股票过滤: 过滤 /y_2017, 3947/0/4753

[2022-06-25 03:07:03.341519] INFO: A股股票过滤: 过滤 /y_2018, 2772/0/3562

[2022-06-25 03:07:04.906248] INFO: A股股票过滤: 过滤 /y_2019, 3495/0/4677

[2022-06-25 03:07:04.912023] INFO: A股股票过滤: 过滤完成, 29473 + 0

[2022-06-25 03:07:04.954081] INFO: moduleinvoker: chinaa_stock_filter.v1 运行完成[9.146729s].

[2022-06-25 03:07:04.967022] INFO: moduleinvoker: dropnan.v1 开始运行..

[2022-06-25 03:07:05.146078] INFO: dropnan: /y_2013, 1556/1561

[2022-06-25 03:07:05.208444] INFO: dropnan: /y_2014, 2873/2873

[2022-06-25 03:07:05.285622] INFO: dropnan: /y_2015, 10005/10010

[2022-06-25 03:07:05.365050] INFO: dropnan: /y_2016, 4808/4815

[2022-06-25 03:07:05.497679] INFO: dropnan: /y_2017, 3946/3947

[2022-06-25 03:07:05.563342] INFO: dropnan: /y_2018, 2765/2772

[2022-06-25 03:07:05.627030] INFO: dropnan: /y_2019, 3494/3495

[2022-06-25 03:07:05.692005] INFO: dropnan: 行数: 29447/29473

[2022-06-25 03:07:05.774498] INFO: moduleinvoker: dropnan.v1 运行完成[0.807475s].

[2022-06-25 03:07:05.781260] INFO: moduleinvoker: sort.v5 开始运行..

[2022-06-25 03:07:10.452250] INFO: moduleinvoker: sort.v5 运行完成[4.670976s].

[2022-06-25 03:07:10.461590] INFO: moduleinvoker: stock_ranker_train.v6 开始运行..

[2022-06-25 03:07:10.633093] INFO: StockRanker: 特征预处理 ..

[2022-06-25 03:07:10.722193] INFO: StockRanker: prepare data: training ..

[2022-06-25 03:07:11.127762] INFO: StockRanker训练: d948e89a 准备训练: 29447 行数

[2022-06-25 03:07:11.130005] INFO: StockRanker训练: AI模型训练，将在29447*11=32.39万数据上对模型训练进行20轮迭代训练。预计将需要1～2分钟。请耐心等待。

[2022-06-25 03:07:11.326843] INFO: StockRanker训练: 正在训练 ..

[2022-06-25 03:07:11.377912] INFO: StockRanker训练: 任务状态: Pending

[2022-06-25 03:07:21.490046] INFO: StockRanker训练: 任务状态: Running

[2022-06-25 03:08:21.782211] INFO: StockRanker训练: 任务状态: Succeeded

[2022-06-25 03:08:21.795166] ERROR: moduleinvoker: module name: stock_ranker_train, module version: v6, trackeback: Exception: 模型训练失败：可能导致错误的原因是训练数据问题，请检查训练数据, err_code=1 (d948e89af3f011ec91d11ebacbcb8fc8)

---------------------------------------------------------------------------
Exception                                 Traceback (most recent call last)
<ipython-input-3-7ba11a0d804d> in <module>
    195 )
    196 
--> 197 m6 = M.stock_ranker_train.v6(
    198     training_ds=m11.sorted_data,
    199     features=m3.data,

Exception: 模型训练失败：可能导致错误的原因是训练数据问题，请检查训练数据, err_code=1 (d948e89af3f011ec91d11ebacbcb8fc8)