克隆策略

In [ ]:

# 本代码由可视化策略环境自动生成 2021年9月13日 10:10
# 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 m26_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 = 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 = 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 = 1
    context.options['hold_days'] = 1
# 回测引擎：每日数据处理函数，每天执行一次
def m26_handle_data_bigquant_run(context, data):
    #获取当日日期
    today_date = data.current_dt.strftime('%Y-%m-%d')
    
    #大盘风控模块，读取风控数据    
    benckmark_risk=context.benckmark_risk.loc[today_date].values[0]

    #当risk为1时，市场有风险，全部平仓，不再执行其它操作
    if benckmark_risk > 0:
        position_all = context.portfolio.positions.keys()
        for i in position_all:
            context.order_target(i, 0)
        print(today_date,'大盘风控止损触发,全仓卖出')
        return
    
    # 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) * 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()}
    
    #--------------------------START：持有固定天数卖出(不含建仓期)---------------
    current_stopdays_stock = [] 
    today = data.current_dt
    today_date = data.current_dt.strftime('%Y-%m-%d')
    # 不是建仓期（在前hold_days属于建仓期）
    if not is_staging:
        equities = {e.symbol: p for e, p in context.portfolio.positions.items() if p.amount>0}
        if len(equities)>0:
            for i in equities:
                sid = equities[i].sid  # 交易标的
                stock_cost=equities[i].cost_basis # 成本价
                stock_market_price = data.current(context.symbol(i), 'price')  # 最新市场价格
                last_sale_date = equities[i].last_sale_date   # 上次交易日期
                delta_days = data.current_dt - last_sale_date
                hold_days = delta_days.days # 持仓天数
                total_of_profit = stock_market_price/stock_cost - 1 # 持仓收益
                highest_price_since_buy = data.history(context.symbol(i), 'close', hold_days, '1d').max() # 建仓以来的最高价
                # 日均收益小于 9.7% 就全部卖出
                if total_of_profit/hold_days < 0.97:
                    context.order_target_percent(sid, 0)
                    current_stopdays_stock.append(i)
                # 当日下跌 直接卖出
                if i not in current_stopdays_stock and highest_price_since_buy and highest_price_since_buy > stock_market_price:
                    context.order_target_percent(sid, 0)
                    current_stopdays_stock.append(i)
            #if len(current_stopdays_stock)>0:        
                #print(today_date,'日均收益小于9.7% 就全部卖出',current_stopdays_stock)
    #-------------------------------END:持有固定天数卖出--------------------------    
    
    
    # 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]))])))
        for instrument in instruments:
            #防止多个止损条件同时满足，出现多次卖出产生空单
            if instrument not in current_stopdays_stock:
                context.order_target(context.symbol(instrument), 0)
                cash_for_sell -= positions[instrument]
            else:
                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_tmp = list(ranker_prediction.instrument)
    #防止卖出后再次买入
    buy_instruments=[k for k in buy_instruments_tmp if k not in current_stopdays_stock][: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:
            current_price = data.current(context.symbol(instrument), 'price')
            amount = math.floor(cash / current_price / 100) * 100
            context.order(context.symbol(instrument), amount)

# 回测引擎：准备数据，只执行一次
def m26_prepare_bigquant_run(context):
    #在数据准备函数中一次性计算每日的大盘风控条件相比于在handle中每日计算风控条件可以提高回测速度
    # 多取50天的数据便于计算均值(保证回测的第一天均值不为Nan值)，其中context.start_date和context.end_date是回测指定的起始时间和终止时间
    start_date= (pd.to_datetime(context.start_date) - datetime.timedelta(days=50)).strftime('%Y-%m-%d')     
    benckmark_data=D.history_data(instruments=['000001.SZA'], start_date=start_date, end_date=context.end_date,fields=['close'])
    #计算指数5日涨幅
    benckmark_data['ret5']=benckmark_data['close']/benckmark_data['close'].shift(5)-1
    #计算大盘风控条件，如果5日涨幅小于-10%则设置风险状态risk为1，否则为0
    benckmark_data['risk'] = np.where(benckmark_data['ret5']<-0.1,0,0)
    #修改日期格式为字符串(便于在handle中使用字符串日期索引来查看每日的风险状态)
    benckmark_data['date']=benckmark_data['date'].apply(lambda x:x.strftime('%Y-%m-%d'))
    #设置日期为索引
    benckmark_data.set_index('date',inplace=True)
    #把风控序列输出给全局变量context.benckmark_risk
    context.benckmark_risk=benckmark_data[['risk']]

m1 = M.instruments.v2(
    start_date='2010-01-01',
    end_date='2019-12-31',
    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 {{web_host_url}}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 <{{web_host_url}}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, -2) / 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
)

m14 = M.standardlize.v8(
    input_1=m2.data,
    columns_input='label'
)

m3 = M.input_features.v1(
    features="""#std(deal_number_0, 3)
std(turn_0, 3)
#std(volume_0, 3)
#std(amount_0, 3)
#std(return_0, 3)
zf1 = 1*abs((close_0 - open_0)/close_1)
close_0/mean(close_0, 5)
#(high_0 - low_0)/close_1
zf2 = (high_2 - low_2)/close_3 + (high_1 - low_1)/close_2 + (high_0 - low_0)/close_1
close_0/open_0
close_0/high_0
close_0/((high_0 + low_0)/2)
zf3 = (close_0 - ts_min(close_0,5))/ts_min(close_0,5)
#zf4 = (close_0 - ts_min(close_0,10))/ts_min(close_0,10)
#zf5 = (close_0 - ts_min(close_0,20))/ts_min(close_0,20)"""
)

m15 = M.general_feature_extractor.v7(
    instruments=m1.data,
    features=m3.data,
    start_date='',
    end_date='',
    before_start_days=250
)

m4 = M.chinaa_stock_filter.v1(
    input_data=m15.data,
    index_constituent_cond=['全部'],
    board_cond=['上证主板', '深证主板', '创业板'],
    industry_cond=['全部'],
    st_cond=['正常'],
    delist_cond=['非退市'],
    output_left_data=False
)

m16 = M.derived_feature_extractor.v3(
    input_data=m4.data,
    features=m3.data,
    date_col='date',
    instrument_col='instrument',
    drop_na=False,
    remove_extra_columns=False
)

m9 = M.instruments.v2(
    start_date=T.live_run_param('trading_date', '2020-01-01'),
    end_date=T.live_run_param('trading_date', '2021-08-31'),
    market='CN_STOCK_A',
    instrument_list='',
    max_count=0
)

m18 = M.general_feature_extractor.v7(
    instruments=m9.data,
    features=m3.data,
    start_date='',
    end_date='',
    before_start_days=250
)

m11 = M.chinaa_stock_filter.v1(
    input_data=m18.data,
    index_constituent_cond=['全部'],
    board_cond=['上证主板', '深证主板', '创业板'],
    industry_cond=['全部'],
    st_cond=['正常'],
    delist_cond=['非退市'],
    output_left_data=False
)

m19 = M.derived_feature_extractor.v3(
    input_data=m11.data,
    features=m3.data,
    date_col='date',
    instrument_col='instrument',
    drop_na=False,
    remove_extra_columns=False
)

m5 = M.input_features.v1(
    features="""yz_1 = where(mean(close_0, 5) >= mean(close_0, 10), 1, 0)
yz_2 = where(mean(close_0, 10) >= mean(close_0, 20), 1, 0)
yz_3 = where(close_0 >= mean(close_0, 5), 1, 0)
yz_4 = where(high_0 >= ts_max(high_0, 5), 1, 0)
yz_5 = where(zf1 <= 0.04, 1, 0)
yz_6 = where((close_0 - ts_min(close_0,20))/ts_min(close_0,20) < 0.4, 1, 0)"""
)

m10 = M.derived_feature_extractor.v3(
    input_data=m19.data,
    features=m5.data,
    date_col='date',
    instrument_col='instrument',
    drop_na=False,
    remove_extra_columns=False,
    user_functions={}
)

m12 = M.filter.v3(
    input_data=m10.data,
    expr='yz_1 == True and yz_2 == True and yz_3 == True and yz_4 == True and yz_5 == True and yz_6 == True',
    output_left_data=False
)

m22 = M.dropnan.v2(
    input_data=m12.data
)

m20 = M.input_features.v1(
    features="""yz_1 = where(mean(close_0, 5) >= mean(close_0, 10), 1, 0)
yz_2 = where(mean(close_0, 10) >= mean(close_0, 20), 1, 0)
yz_3 = where(close_0 >= mean(close_0, 5), 1, 0)
yz_4 = where(high_0 >= ts_max(high_0, 5), 1, 0)
yz_5 = where(zf1 <= 0.04, 1, 0)
yz_6 = where((close_0 - ts_min(close_0,20))/ts_min(close_0,20) < 0.4, 1, 0)"""
)

m23 = M.derived_feature_extractor.v3(
    input_data=m16.data,
    features=m20.data,
    date_col='date',
    instrument_col='instrument',
    drop_na=False,
    remove_extra_columns=False,
    user_functions={}
)

m24 = M.filter.v3(
    input_data=m23.data,
    expr='yz_1 == True and yz_2 == True and yz_3 == True and yz_4 == True and yz_5 == True and yz_6 == True',
    output_left_data=False
)

m7 = M.join.v3(
    data1=m14.data,
    data2=m24.data,
    on='date,instrument',
    how='inner',
    sort=False
)

m21 = M.dropnan.v2(
    input_data=m7.data
)

m6 = M.filter.v3(
    input_data=m21.data,
    expr='date < "2018-01-01"',
    output_left_data=True
)

m17 = M.stock_ranker_train.v6(
    training_ds=m6.data,
    features=m3.data,
    test_ds=m6.left_data,
    learning_algorithm='排序',
    number_of_leaves=80,
    minimum_docs_per_leaf=1000,
    number_of_trees=60,
    learning_rate=0.1,
    max_bins=1023,
    feature_fraction=1,
    data_row_fraction=1,
    ndcg_discount_base=1,
    m_lazy_run=False
)

m8 = M.stock_ranker_predict.v5(
    model=m17.model,
    data=m22.data,
    m_lazy_run=False
)

m26 = M.trade.v4(
    instruments=m9.data,
    options_data=m8.predictions,
    start_date='',
    end_date='',
    initialize=m26_initialize_bigquant_run,
    handle_data=m26_handle_data_bigquant_run,
    prepare=m26_prepare_bigquant_run,
    volume_limit=0.025,
    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.SHA'
)

[2021-09-13 10:08:20.774373] INFO: moduleinvoker: instruments.v2 开始运行..

[2021-09-13 10:08:20.784867] INFO: moduleinvoker: 命中缓存

[2021-09-13 10:08:20.786458] INFO: moduleinvoker: instruments.v2 运行完成[0.012098s].

[2021-09-13 10:08:20.793754] INFO: moduleinvoker: advanced_auto_labeler.v2 开始运行..

[2021-09-13 10:08:20.815099] INFO: moduleinvoker: 命中缓存

[2021-09-13 10:08:20.817854] INFO: moduleinvoker: advanced_auto_labeler.v2 运行完成[0.024094s].

[2021-09-13 10:08:20.826554] INFO: moduleinvoker: standardlize.v8 开始运行..

[2021-09-13 10:08:20.833858] INFO: moduleinvoker: 命中缓存

[2021-09-13 10:08:20.835471] INFO: moduleinvoker: standardlize.v8 运行完成[0.008942s].

[2021-09-13 10:08:20.839468] INFO: moduleinvoker: input_features.v1 开始运行..

[2021-09-13 10:08:20.845351] INFO: moduleinvoker: 命中缓存

[2021-09-13 10:08:20.847179] INFO: moduleinvoker: input_features.v1 运行完成[0.007711s].

[2021-09-13 10:08:20.861214] INFO: moduleinvoker: general_feature_extractor.v7 开始运行..

[2021-09-13 10:08:20.868112] INFO: moduleinvoker: 命中缓存

[2021-09-13 10:08:20.869601] INFO: moduleinvoker: general_feature_extractor.v7 运行完成[0.008402s].

[2021-09-13 10:08:20.877459] INFO: moduleinvoker: chinaa_stock_filter.v1 开始运行..

[2021-09-13 10:08:20.887386] INFO: moduleinvoker: 命中缓存

[2021-09-13 10:08:20.888908] INFO: moduleinvoker: chinaa_stock_filter.v1 运行完成[0.011455s].

[2021-09-13 10:08:20.896235] INFO: moduleinvoker: derived_feature_extractor.v3 开始运行..

[2021-09-13 10:08:20.907318] INFO: moduleinvoker: 命中缓存

[2021-09-13 10:08:20.910105] INFO: moduleinvoker: derived_feature_extractor.v3 运行完成[0.013851s].

[2021-09-13 10:08:20.918275] INFO: moduleinvoker: instruments.v2 开始运行..

[2021-09-13 10:08:20.927944] INFO: moduleinvoker: 命中缓存

[2021-09-13 10:08:20.929969] INFO: moduleinvoker: instruments.v2 运行完成[0.011688s].

[2021-09-13 10:08:20.942762] INFO: moduleinvoker: general_feature_extractor.v7 开始运行..

[2021-09-13 10:08:20.953668] INFO: moduleinvoker: 命中缓存

[2021-09-13 10:08:20.955612] INFO: moduleinvoker: general_feature_extractor.v7 运行完成[0.012867s].

[2021-09-13 10:08:20.963557] INFO: moduleinvoker: chinaa_stock_filter.v1 开始运行..

[2021-09-13 10:08:20.979911] INFO: moduleinvoker: 命中缓存

[2021-09-13 10:08:20.981496] INFO: moduleinvoker: chinaa_stock_filter.v1 运行完成[0.017944s].

[2021-09-13 10:08:20.988228] INFO: moduleinvoker: derived_feature_extractor.v3 开始运行..

[2021-09-13 10:08:20.994982] INFO: moduleinvoker: 命中缓存

[2021-09-13 10:08:20.996718] INFO: moduleinvoker: derived_feature_extractor.v3 运行完成[0.008491s].

[2021-09-13 10:08:21.001533] INFO: moduleinvoker: input_features.v1 开始运行..

[2021-09-13 10:08:21.035129] INFO: moduleinvoker: input_features.v1 运行完成[0.033597s].

[2021-09-13 10:08:21.043765] INFO: moduleinvoker: derived_feature_extractor.v3 开始运行..

[2021-09-13 10:08:31.680773] INFO: derived_feature_extractor: 提取完成 yz_1 = where(mean(close_0, 5) >= mean(close_0, 10), 1, 0), 3.150s

[2021-09-13 10:08:34.705716] INFO: derived_feature_extractor: 提取完成 yz_2 = where(mean(close_0, 10) >= mean(close_0, 20), 1, 0), 3.023s

[2021-09-13 10:08:36.180372] INFO: derived_feature_extractor: 提取完成 yz_3 = where(close_0 >= mean(close_0, 5), 1, 0), 1.472s

[2021-09-13 10:08:37.734895] INFO: derived_feature_extractor: 提取完成 yz_4 = where(high_0 >= ts_max(high_0, 5), 1, 0), 1.553s

[2021-09-13 10:08:37.741503] INFO: derived_feature_extractor: 提取完成 yz_5 = where(zf1 <= 0.04, 1, 0), 0.005s

[2021-09-13 10:08:40.847778] INFO: derived_feature_extractor: 提取完成 yz_6 = where((close_0 - ts_min(close_0,20))/ts_min(close_0,20) < 0.4, 1, 0), 3.105s

[2021-09-13 10:08:42.977957] INFO: derived_feature_extractor: /y_2019, 587293

[2021-09-13 10:08:46.602675] INFO: derived_feature_extractor: /y_2020, 870321

[2021-09-13 10:08:49.787460] INFO: derived_feature_extractor: /y_2021, 617504

[2021-09-13 10:08:50.566400] INFO: moduleinvoker: derived_feature_extractor.v3 运行完成[29.522619s].

[2021-09-13 10:08:50.575345] INFO: moduleinvoker: filter.v3 开始运行..

[2021-09-13 10:08:50.588740] INFO: filter: 使用表达式 yz_1 == True and yz_2 == True and yz_3 == True and yz_4 == True and yz_5 == True and yz_6 == True 过滤

[2021-09-13 10:08:51.219905] INFO: filter: 过滤 /y_2019, 43772/0/587293

[2021-09-13 10:08:52.114611] INFO: filter: 过滤 /y_2020, 64424/0/870321

[2021-09-13 10:08:52.762845] INFO: filter: 过滤 /y_2021, 41867/0/617504

[2021-09-13 10:08:52.798635] INFO: moduleinvoker: filter.v3 运行完成[2.223286s].

[2021-09-13 10:08:52.807072] INFO: moduleinvoker: dropnan.v2 开始运行..

[2021-09-13 10:08:52.965965] INFO: dropnan: /y_2019, 43772/43772

[2021-09-13 10:08:53.090244] INFO: dropnan: /y_2020, 64424/64424

[2021-09-13 10:08:53.188812] INFO: dropnan: /y_2021, 41867/41867

[2021-09-13 10:08:53.295360] INFO: dropnan: 行数: 150063/150063

[2021-09-13 10:08:53.301689] INFO: moduleinvoker: dropnan.v2 运行完成[0.494608s].

[2021-09-13 10:08:53.312264] INFO: moduleinvoker: input_features.v1 开始运行..

[2021-09-13 10:08:53.319126] INFO: moduleinvoker: 命中缓存

[2021-09-13 10:08:53.320567] INFO: moduleinvoker: input_features.v1 运行完成[0.008464s].

[2021-09-13 10:08:53.327885] INFO: moduleinvoker: derived_feature_extractor.v3 开始运行..

[2021-09-13 10:09:27.155371] INFO: derived_feature_extractor: 提取完成 yz_1 = where(mean(close_0, 5) >= mean(close_0, 10), 1, 0), 10.490s

[2021-09-13 10:09:38.585734] INFO: derived_feature_extractor: 提取完成 yz_2 = where(mean(close_0, 10) >= mean(close_0, 20), 1, 0), 11.429s

[2021-09-13 10:09:44.290514] INFO: derived_feature_extractor: 提取完成 yz_3 = where(close_0 >= mean(close_0, 5), 1, 0), 5.703s

[2021-09-13 10:09:50.151247] INFO: derived_feature_extractor: 提取完成 yz_4 = where(high_0 >= ts_max(high_0, 5), 1, 0), 5.859s

[2021-09-13 10:09:50.163640] INFO: derived_feature_extractor: 提取完成 yz_5 = where(zf1 <= 0.04, 1, 0), 0.010s

[2021-09-13 10:10:01.564237] INFO: derived_feature_extractor: 提取完成 yz_6 = where((close_0 - ts_min(close_0,20))/ts_min(close_0,20) < 0.4, 1, 0), 11.399s

[2021-09-13 10:10:02.854853] INFO: derived_feature_extractor: /y_2009, 249804

[2021-09-13 10:10:04.750592] INFO: derived_feature_extractor: /y_2010, 413881

[2021-09-13 10:10:07.045133] INFO: derived_feature_extractor: /y_2011, 493517

[2021-09-13 10:10:09.515847] INFO: derived_feature_extractor: /y_2012, 548410

[2021-09-13 10:10:12.074672] INFO: derived_feature_extractor: /y_2013, 546947

[2021-09-13 10:10:14.740904] INFO: derived_feature_extractor: /y_2014, 554064

[2021-09-13 10:10:17.433839] INFO: derived_feature_extractor: /y_2015, 556950