复制链接

克隆策略

In [ ]:

# 本代码由可视化策略环境自动生成 2022年3月26日 01:16
# 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):
    print('初始化函数，只执行一次')
    # 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 = 10
    # 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 = 0.3
    context.options['hold_days'] = 10

# 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
    ranker_prediction = context.ranker_prediction[
        context.ranker_prediction.date == data.current_dt.strftime('%Y-%m-%d')]
    
    today = data.current_dt.strftime('%Y-%m-%d')
    #print('日期:',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']
    #print('context.portfolio.portfolio_value:',context.portfolio.portfolio_value)
    cash_for_buy = min(context.portfolio.cash, (1 if is_staging else 1.5) * cash_avg)
    #print('cash_for_buy:',cash_for_buy,' context.portfolio.cash:',context.portfolio.cash)
    cash_for_sell = cash_avg - (context.portfolio.cash - cash_for_buy)
    #print('context.portfolio.portfolio_value:',context.portfolio.portfolio_value)
    positions = {e.symbol: p.amount * p.last_sale_price
                 for e, p in context.portfolio.positions.items()}
    #print('is_staging:',is_staging,' cash_for_sell:',cash_for_sell)

    # 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: p for e, p in context.portfolio.positions.items() if p.amount>0}
                
        instruments = list(reversed(list(ranker_prediction.instrument[ranker_prediction.instrument.apply(
                lambda x: x in equities)])))

        print(today,' 选股: ',ranker_prediction[:10])
        print('sell instruments:',instruments)
        for instrument in instruments:
            context.order_target(context.symbol(instrument), 0)
            cash_for_sell -= positions[instrument]
            
            stock_market_today_high = data.current(context.symbol(instrument), 'high') #今日最高价      
            stock_market_today_close = data.current(context.symbol(instrument), 'close') #今日收盘价
            last_cost_price = equities[instrument].cost_basis # 上次交易金额   
            target_return = stock_market_today_close/last_cost_price

            
            print(today,' instrument 滚动卖出 :','收益： ',target_return, instrument,'context.symbol(instrument):',context.symbol(instrument))
            if cash_for_sell <= 0:
                break
                
         #加上持仓超过50天或者收益大于20%卖出
        if len(equities) > 0:
            for i in equities.keys():
               #print(today,' equities:',equities)
                #p = context.portfolio.positions.items(i)
                #cash_for_sell -= p.amount * p.last_sale_price
                
                stock_market_price = data.current(context.symbol(i), 'price')  # 最新市场价格
                stock_market_today_high = data.current(context.symbol(i), 'high') #今日最高价      
                stock_market_today_close = data.current(context.symbol(i), 'close') #今日收盘价
                last_sale_date = equities[i].last_sale_date   # 上次交易日期
                last_cost_price = equities[i].cost_basis # 上次交易金额
                delta_days = data.current_dt - last_sale_date  
                hold_days = delta_days.days # 持仓天数
        # 最高收益
        #high_return = (highclose_price_since_buy-last_cost_price)/last_cost_price
        
                target_return = stock_market_today_close/last_cost_price
               
                if hold_days>=100 :
                    context.order_target(context.symbol(i), 0)
                    print(today,'超期卖出 :','收益： ',target_return,equities[i], ' context.symbol(i):',context.symbol(i))
                    context.order_target(context.symbol(i), 0)
                #if target_return>=1.2 :
                #    context.order_target(context.symbol(i), 0)
                #    print(today,' 盈利卖出 :','收益： ',target_return,equities[i],' context.symbol(i):',context.symbol(i))
               # elif target_return<=0.9 :
               #     context.order_target(context.symbol(i), 0)
               #     print(today,' 止损卖出 :','收益： ',target_return,equities[i],' context.symbol(i):',context.symbol(i))
                

    # 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)])
    print(today,' buy_instruments:',buy_instruments,'  权重： ',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)
            print(today,' 买入 ',instrument)

# 回测引擎：准备数据，只执行一次
def m19_prepare_bigquant_run(context):
    # 加载预测数据
    print('准备数据，只执行一次')
    df = context.options['data'].read_df()
    # 函数：求满足开仓条件的股票列表
    def open_pos_con(df):
        return list(df[df['fantanbili']>0].instrument)
    # 函数：求满足平仓条件的股票列表
    def close_pos_con(df):
        return list(df[df['fantanbili']>0].instrument)
    
    # 每日卖出股票的数据框
    context.daily_sell_stock= df.groupby('date').apply(close_pos_con)  
    # 每日买入股票的数据框
    context.daily_buy_stock= df.groupby('date').apply(open_pos_con)  


m1 = M.instruments.v2(
    start_date='2019-01-01',
    end_date='2019-06-30',
    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, -100) / 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.HIX',
    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
#pe_ttm_0

#high_0/adjust_factor_0
#low_0/adjust_factor_0

#timeperiod移动平均线
#zhouma34=ta_ma(close_0, timeperiod=166)/adjust_factor_0
zhouma34=ta_ma(close_0, timeperiod=166)
#timeperiod移动平均线
zhouma55=ta_ma(close_0, timeperiod=258)
isup=zhouma34/zhouma55
sy=shift(close_0, -100)/shift(open_0, -1)

#ts_max(high_0,10)/adjust_factor_0
#ts_min(low_0,10)/adjust_factor_0

ts_argmax(high_0, 258*5)
#4年高点，1年按258个交易日算
isgaodian=where(ts_argmax(high_0, 258*5)<500.0,1,0)

#1年低点，1年按258个交易日算
ts_argmin(low_0, 258*1)
isdidian=where(ts_argmin(low_0, 258*1)<100.0,1,0)

#半年高点，1年按258个交易日算
isbanniangaodian=where(ts_argmax(high_0, 128)<80.0,1,0)
#在周34-55均线区间，日166-258
iszaiquejian=where((close_0<zhouma34) & (close_0>zhouma55),1,0)
#曾经前几天下过55线，触碰过日258线
isdiyu258=where(ts_min(low_0, 3)<zhouma55,1,0)

#tmax=ts_argmax(high_0, 258*5)

#反弹比例,选比例最高的10-20只交易，后面调试
yiniandidian=ts_min(low_0, 258*1)
banniangaodian=ts_max(high_0, 128*1)
fantanbili=banniangaodian/yiniandidian

#排除ST
st_status_0
#时间序列函数， d 天内的最大值
#ts_max(high_0, 258*4)
#时间序列函数， d 天内的最小值
#ts_min(low_0, 258*1)



#isxiadie=where(ts_max(high_0, 258*4)>ts_min(low_0, 258*1),1,0)
"""
)

m15 = M.general_feature_extractor.v7(
    instruments=m1.data,
    features=m3.data,
    start_date='',
    end_date='',
    before_start_days=2500
)

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

m5 = M.filter.v3(
    input_data=m16.data,
    expr='isgaodian==1&isdidian==1&isbanniangaodian==1&st_status_0==0&date>\'2019-01-01\'',
    output_left_data=True
)

m10 = M.sort.v5(
    input_ds=m5.data,
    sort_by='isup',
    group_by='date',
    keep_columns='--',
    ascending=False
)

m7 = M.join.v3(
    data1=m2.data,
    data2=m10.sorted_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=30,
    minimum_docs_per_leaf=1000,
    number_of_trees=20,
    learning_rate=0.1,
    max_bins=1023,
    feature_fraction=1,
    m_lazy_run=False
)

m9 = M.instruments.v2(
    start_date=T.live_run_param('trading_date', '2019-07-01'),
    end_date=T.live_run_param('trading_date', '2019-12-30'),
    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=2500
)

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

m12 = M.filter.v3(
    input_data=m18.data,
    expr='isgaodian==1&isdidian==1&isbanniangaodian==1&st_status_0==0&date>\'2019-01-01\'',
    output_left_data=True
)

m20 = M.sort.v5(
    input_ds=m12.data,
    sort_by='isup',
    group_by='date',
    keep_columns='--',
    ascending=False
)

m14 = M.dropnan.v1(
    input_data=m20.sorted_data
)

m8 = M.stock_ranker_predict.v5(
    model=m6.model,
    data=m14.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.HIX'
)

[2022-03-26 01:16:24.287076] INFO: moduleinvoker: instruments.v2 开始运行..

[2022-03-26 01:16:24.295721] INFO: moduleinvoker: 命中缓存

[2022-03-26 01:16:24.297321] INFO: moduleinvoker: instruments.v2 运行完成[0.010229s].

[2022-03-26 01:16:24.310654] INFO: moduleinvoker: advanced_auto_labeler.v2 开始运行..

[2022-03-26 01:16:24.318261] INFO: moduleinvoker: 命中缓存

[2022-03-26 01:16:24.319922] INFO: moduleinvoker: advanced_auto_labeler.v2 运行完成[0.009268s].

[2022-03-26 01:16:24.325837] INFO: moduleinvoker: input_features.v1 开始运行..

[2022-03-26 01:16:24.333887] INFO: moduleinvoker: 命中缓存

[2022-03-26 01:16:24.335287] INFO: moduleinvoker: input_features.v1 运行完成[0.009446s].

[2022-03-26 01:16:24.348790] INFO: moduleinvoker: general_feature_extractor.v7 开始运行..

[2022-03-26 01:16:24.355214] INFO: moduleinvoker: 命中缓存

[2022-03-26 01:16:24.356573] INFO: moduleinvoker: general_feature_extractor.v7 运行完成[0.007785s].

[2022-03-26 01:16:24.363085] INFO: moduleinvoker: derived_feature_extractor.v3 开始运行..

[2022-03-26 01:16:24.369257] INFO: moduleinvoker: 命中缓存

[2022-03-26 01:16:24.370645] INFO: moduleinvoker: derived_feature_extractor.v3 运行完成[0.007557s].

[2022-03-26 01:16:24.378253] INFO: moduleinvoker: filter.v3 开始运行..

[2022-03-26 01:16:24.386591] INFO: moduleinvoker: 命中缓存

[2022-03-26 01:16:24.388030] INFO: moduleinvoker: filter.v3 运行完成[0.009777s].

[2022-03-26 01:16:24.393389] INFO: moduleinvoker: sort.v5 开始运行..

[2022-03-26 01:16:24.401011] INFO: moduleinvoker: 命中缓存

[2022-03-26 01:16:24.402293] INFO: moduleinvoker: sort.v5 运行完成[0.008901s].

[2022-03-26 01:16:24.409382] INFO: moduleinvoker: join.v3 开始运行..

[2022-03-26 01:16:24.416086] INFO: moduleinvoker: 命中缓存

[2022-03-26 01:16:24.417259] INFO: moduleinvoker: join.v3 运行完成[0.007875s].

[2022-03-26 01:16:24.424938] INFO: moduleinvoker: dropnan.v1 开始运行..

[2022-03-26 01:16:24.431578] INFO: moduleinvoker: 命中缓存

[2022-03-26 01:16:24.432798] INFO: moduleinvoker: dropnan.v1 运行完成[0.007857s].

[2022-03-26 01:16:24.445518] INFO: moduleinvoker: stock_ranker_train.v5 开始运行..

[2022-03-26 01:16:24.532492] INFO: StockRanker: 特征预处理 ..

[2022-03-26 01:16:24.559297] INFO: StockRanker: prepare data: training ..

[2022-03-26 01:16:24.621523] INFO: StockRanker训练: 4c5b351a 准备训练: 930 行数

[2022-03-26 01:16:24.842735] INFO: StockRanker训练: 正在训练 ..

In [48]:

m10.sorted_data.read_df()
m2.plot_label_counts()