克隆策略

深度学习在期货高频上的应用¶

策略思想：¶

使用最近50分钟的价格、成交量、持仓量数据预测未来50分钟的涨跌幅。

交易标的：¶

股指期货 IF1906

模型算法：¶

LSTM深度学习算法

模型标注：¶

未来50分钟收益率

模型因子：¶

mean(open_intl,5)/amount
mean(open_intl,10)/amount
mean(open_intl,20)/amount
mean(open_intl,30)/amount
mean(open_intl,50)/amount
close/shift(close,30)
close/shift(close,20)
close/shift(close,10)
close/shift(close,5)
mean(amount,30)
max(high,10)/close 
open_intl
sum(open_intl,10)/amount 
amount/open_intl
mean(amount/open_intl,5)
mean(amount/open_intl,10)
mean(open_intl,5)

交易信号：¶

当预测值大于0.2时并且没有多仓，进场做多。（如果有空单，需要先平掉空单）
当预测值小于-0.2时并且没有空仓，进场做空。（如果有多单，需要先平掉多单）

In [4]:

# 本代码由可视化策略环境自动生成 2021年8月19日 09:14
# 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 m20_run_bigquant_run(input_1, input_2, input_3):
     
    start_date=input_1.read_pickle()['start_date']
    end_date=input_1.read_pickle()['end_date']
    ins=input_1.read_pickle()['instruments']
    df = DataSource('bar1m_IF1906.CFE').read(instruments=ins,start_date=start_date,end_date=end_date)
    df['adjust_factor']=1.0
    data_1 = DataSource.write_df(df)
    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 m20_post_run_bigquant_run(outputs):
    return outputs

# 用户的自定义层需要写到字典中，比如
# {
#   "MyLayer": MyLayer
# }
m6_custom_objects_bigquant_run = {
    
}

import pandas as pd

# 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 m2_run_bigquant_run(input_1, input_2, input_3):
    predictions = input_1.read_pickle()
    pred_result = predictions.reshape(predictions.shape[0]) 
    dt = input_2.read_df()['date']
    pred_df = pd.Series(pred_result, index=dt)
    pred_df = pd.DataFrame(pred_df)
    ds = DataSource.write_df(pred_df)
    return Outputs(data_1=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 m2_post_run_bigquant_run(outputs):
    return outputs

# 交易引擎：初始化函数，只执行一次
def m19_initialize_bigquant_run(context):
    # 加载预测数据
    print("in initial", context.future_symbol("IF1906.CFX"))
    context.prediction = context.options['data'].read().reset_index().rename(columns={0:"value"})

# 交易引擎：每个单位时间开盘前调用一次。
def m19_before_trading_start_bigquant_run(context, data):
    pass

# 交易引擎：bar数据处理函数，每个单位执行一次
def m19_handle_tick_bigquant_run(context, data):
    pass

# 交易引擎：bar数据处理函数，每个单位执行一次
def m19_handle_data_bigquant_run(context, data):
    # 按日期过滤得到今日的预测数据
    now = data.current_dt.strftime('%Y-%m-%d %H:%M:%S')

    # 按日期过滤得到今日的预测数据
    try:
        prediction = context.prediction[context.prediction.date==now].iloc[0]["value"]
    except Exception as e:
        return
    instrument = context.instruments[0]
#     sid = context.symbol(instrument)
#     cur_position = context.portfolio.positions[sid].amount
    # 分别获取多头持仓，和空头持仓
    position_long = context.get_position(instrument, Direction.LONG)
#     position_short = context.get_position(instrument, Direction.SHORT)
    #获取当前最新价
    price = data.current(instrument, "close")
    order_num = 1
    # 交易逻辑
    if prediction > 0.2 and position_long.avail_qty == 0:
        context.buy_open(instrument, order_num, price, order_type=OrderType.MARKET)
        print(data.current_dt, '开多！')
        
    elif prediction < -0.2 and position_long.avail_qty > 0:
        context.sell_close(instrument, position_long.avail_qty, price, order_type=OrderType.MARKET)
        print(data.current_dt, '平多！')

# 交易引擎：成交回报处理函数，每个成交发生时执行一次
def m19_handle_trade_bigquant_run(context, data):
    pass

# 交易引擎：委托回报处理函数，每个委托变化时执行一次
def m19_handle_order_bigquant_run(context, data):
    pass

# 交易引擎：盘后处理函数，每日盘后执行一次
def m19_after_trading_bigquant_run(context, data):
    pass


m3 = M.dl_layer_input.v1(
    shape='50,17',
    batch_shape='',
    dtype='float32',
    sparse=False,
    name=''
)

m4 = M.dl_layer_lstm.v1(
    inputs=m3.data,
    units=32,
    activation='linear',
    recurrent_activation='hard_sigmoid',
    use_bias=True,
    kernel_initializer='glorot_uniform',
    recurrent_initializer='Orthogonal',
    bias_initializer='Ones',
    unit_forget_bias=True,
    kernel_regularizer='None',
    kernel_regularizer_l1=0,
    kernel_regularizer_l2=0,
    recurrent_regularizer='None',
    recurrent_regularizer_l1=0,
    recurrent_regularizer_l2=0,
    bias_regularizer='None',
    bias_regularizer_l1=0,
    bias_regularizer_l2=0,
    activity_regularizer='None',
    activity_regularizer_l1=0,
    activity_regularizer_l2=0,
    kernel_constraint='None',
    recurrent_constraint='None',
    bias_constraint='None',
    dropout=0,
    recurrent_dropout=0,
    return_sequences=True,
    implementation='0',
    name=''
)

m10 = M.dl_layer_dropout.v1(
    inputs=m4.data,
    rate=0.2,
    noise_shape='',
    seed=0,
    name=''
)

m25 = M.dl_layer_lstm.v1(
    inputs=m10.data,
    units=32,
    activation='sigmoid',
    recurrent_activation='hard_sigmoid',
    use_bias=True,
    kernel_initializer='glorot_uniform',
    recurrent_initializer='Orthogonal',
    bias_initializer='Zeros',
    unit_forget_bias=True,
    kernel_regularizer='None',
    kernel_regularizer_l1=0,
    kernel_regularizer_l2=0,
    recurrent_regularizer='None',
    recurrent_regularizer_l1=0,
    recurrent_regularizer_l2=0,
    bias_regularizer='None',
    bias_regularizer_l1=0,
    bias_regularizer_l2=0,
    activity_regularizer='None',
    activity_regularizer_l1=0,
    activity_regularizer_l2=0,
    kernel_constraint='None',
    recurrent_constraint='None',
    bias_constraint='None',
    dropout=0,
    recurrent_dropout=0,
    return_sequences=False,
    implementation='0',
    name=''
)

m11 = M.dl_layer_dropout.v1(
    inputs=m25.data,
    rate=0.1,
    noise_shape='',
    seed=0,
    name=''
)

m9 = M.dl_layer_dense.v1(
    inputs=m11.data,
    units=1,
    activation='linear',
    use_bias=True,
    kernel_initializer='glorot_uniform',
    bias_initializer='Zeros',
    kernel_regularizer='None',
    kernel_regularizer_l1=0,
    kernel_regularizer_l2=0,
    bias_regularizer='None',
    bias_regularizer_l1=0,
    bias_regularizer_l2=0,
    activity_regularizer='None',
    activity_regularizer_l1=0,
    activity_regularizer_l2=0,
    kernel_constraint='None',
    bias_constraint='None',
    name=''
)

m5 = M.dl_model_init.v1(
    inputs=m3.data,
    outputs=m9.data
)

m8 = M.input_features.v1(
    features="""mean(open_intl,5)/amount
mean(open_intl,10)/amount
mean(open_intl,20)/amount
mean(open_intl,30)/amount
mean(open_intl,50)/amount
close/shift(close,30)
close/shift(close,20)
close/shift(close,10)
close/shift(close,5)
mean(amount,30)
max(high,10)/close 
open_intl
sum(open_intl,10)/amount 
amount/open_intl
mean(amount/open_intl,5)
mean(amount/open_intl,10)
mean(open_intl,5)
label= shift(close,-50)/close-1


"""
)

m24 = M.instruments.v2(
    start_date='2019-04-01',
    end_date='2019-06-22',
    market='CN_FUTURE',
    instrument_list="""IF1906.CFE
 """,
    max_count=0
)

m20 = M.cached.v3(
    input_1=m24.data,
    run=m20_run_bigquant_run,
    post_run=m20_post_run_bigquant_run,
    input_ports='',
    params='{}',
    output_ports=''
)

m12 = M.derived_feature_extractor.v3(
    input_data=m20.data_1,
    features=m8.data,
    date_col='date',
    instrument_col='instrument',
    drop_na=False,
    remove_extra_columns=False,
    user_functions={}
)

m16 = M.filter.v3(
    input_data=m12.data,
    expr='date<\'2019-06-01\'',
    output_left_data=False
)

m28 = M.dropnan.v1(
    input_data=m16.data
)

m21 = M.filter.v3(
    input_data=m12.data,
    expr='date>=\'2019-06-01\'',
    output_left_data=False
)

m13 = M.dropnan.v1(
    input_data=m21.data
)

m15 = M.input_features.v1(
    features="""mean(open_intl,5)/amount
mean(open_intl,10)/amount
mean(open_intl,20)/amount
mean(open_intl,30)/amount
mean(open_intl,50)/amount
close/shift(close,30)
close/shift(close,20)
close/shift(close,10)
close/shift(close,5)
mean(amount,30)
max(high,10)/close 
open_intl
sum(open_intl,10)/amount 
amount/open_intl
mean(amount/open_intl,5)
mean(amount/open_intl,10)
mean(open_intl,5)"""
)

m17 = M.dl_convert_to_bin.v2(
    input_data=m13.data,
    features=m15.data,
    window_size=50,
    feature_clip=5,
    flatten=False,
    window_along_col='instrument'
)

m14 = M.dl_convert_to_bin.v2(
    input_data=m28.data,
    features=m15.data,
    window_size=50,
    feature_clip=5,
    flatten=False,
    window_along_col='instrument'
)

m6 = M.dl_model_train.v1(
    input_model=m5.data,
    training_data=m14.data,
    optimizer='RMSprop',
    loss='mean_squared_error',
    metrics='mse',
    batch_size=256,
    epochs=5,
    custom_objects=m6_custom_objects_bigquant_run,
    n_gpus=0,
    verbose='1:输出进度条记录'
)

m7 = M.dl_model_predict.v1(
    trained_model=m6.data,
    input_data=m17.data,
    batch_size=128,
    n_gpus=0,
    verbose='0:不显示'
)

m2 = M.cached.v3(
    input_1=m7.data,
    input_2=m13.data,
    run=m2_run_bigquant_run,
    post_run=m2_post_run_bigquant_run,
    input_ports='',
    params='{}',
    output_ports=''
)

m18 = M.instruments.v2(
    start_date='2019-06-01',
    end_date='2019-06-22',
    market='CN_FUTURE',
    instrument_list="""IF1906.CFX
 """,
    max_count=0
)

m19 = M.hftrade.v1(
    instruments=m18.data,
    options_data=m2.data_1,
    start_date='',
    end_date='',
    initialize=m19_initialize_bigquant_run,
    before_trading_start=m19_before_trading_start_bigquant_run,
    handle_tick=m19_handle_tick_bigquant_run,
    handle_data=m19_handle_data_bigquant_run,
    handle_trade=m19_handle_trade_bigquant_run,
    handle_order=m19_handle_order_bigquant_run,
    after_trading=m19_after_trading_bigquant_run,
    capital_base=200000,
    frequency='minute',
    price_type='真实价格',
    product_type='期货',
    before_start_days='',
    benchmark='000300.HIX',
    plot_charts=True,
    disable_cache=False,
    show_debug_info=False,
    backtest_only=False
)

[2021-08-18 09:33:10.725418] INFO: moduleinvoker: dl_layer_input.v1 运行完成[0.001557s].

[2021-08-18 09:33:10.850485] INFO: moduleinvoker: dl_layer_lstm.v1 运行完成[0.122188s].

[2021-08-18 09:33:10.856683] INFO: moduleinvoker: dl_layer_dropout.v1 运行完成[0.003259s].

[2021-08-18 09:33:10.979940] INFO: moduleinvoker: dl_layer_lstm.v1 运行完成[0.120503s].

[2021-08-18 09:33:10.985799] INFO: moduleinvoker: dl_layer_dropout.v1 运行完成[0.00305s].

[2021-08-18 09:33:10.997088] INFO: moduleinvoker: dl_layer_dense.v1 运行完成[0.008717s].

[2021-08-18 09:33:11.021622] INFO: moduleinvoker: cached.v3 开始运行..

[2021-08-18 09:33:11.028659] INFO: moduleinvoker: 命中缓存

[2021-08-18 09:33:11.030153] INFO: moduleinvoker: cached.v3 运行完成[0.008546s].

[2021-08-18 09:33:11.031752] INFO: moduleinvoker: dl_model_init.v1 运行完成[0.032277s].

[2021-08-18 09:33:11.033673] INFO: moduleinvoker: input_features.v1 开始运行..

[2021-08-18 09:33:11.039474] INFO: moduleinvoker: 命中缓存

[2021-08-18 09:33:11.040997] INFO: moduleinvoker: input_features.v1 运行完成[0.007324s].

[2021-08-18 09:33:11.043278] INFO: moduleinvoker: instruments.v2 开始运行..

[2021-08-18 09:33:11.048806] INFO: moduleinvoker: 命中缓存

[2021-08-18 09:33:11.050539] INFO: moduleinvoker: instruments.v2 运行完成[0.007252s].

[2021-08-18 09:33:11.055762] INFO: moduleinvoker: cached.v3 开始运行..

[2021-08-18 09:33:11.061863] INFO: moduleinvoker: 命中缓存

[2021-08-18 09:33:11.063237] INFO: moduleinvoker: cached.v3 运行完成[0.007475s].

[2021-08-18 09:33:11.065928] INFO: moduleinvoker: derived_feature_extractor.v3 开始运行..

[2021-08-18 09:33:11.071506] INFO: moduleinvoker: 命中缓存

[2021-08-18 09:33:11.072738] INFO: moduleinvoker: derived_feature_extractor.v3 运行完成[0.006808s].

[2021-08-18 09:33:11.075493] INFO: moduleinvoker: filter.v3 开始运行..

[2021-08-18 09:33:11.080572] INFO: moduleinvoker: 命中缓存

[2021-08-18 09:33:11.081750] INFO: moduleinvoker: filter.v3 运行完成[0.006256s].

[2021-08-18 09:33:11.084761] INFO: moduleinvoker: dropnan.v1 开始运行..

[2021-08-18 09:33:11.090066] INFO: moduleinvoker: 命中缓存

[2021-08-18 09:33:11.091252] INFO: moduleinvoker: dropnan.v1 运行完成[0.006492s].

[2021-08-18 09:33:11.093858] INFO: moduleinvoker: filter.v3 开始运行..

[2021-08-18 09:33:11.099123] INFO: moduleinvoker: 命中缓存

[2021-08-18 09:33:11.100286] INFO: moduleinvoker: filter.v3 运行完成[0.00643s].

[2021-08-18 09:33:11.102831] INFO: moduleinvoker: dropnan.v1 开始运行..

[2021-08-18 09:33:11.107934] INFO: moduleinvoker: 命中缓存

[2021-08-18 09:33:11.109137] INFO: moduleinvoker: dropnan.v1 运行完成[0.006307s].

[2021-08-18 09:33:11.110971] INFO: moduleinvoker: input_features.v1 开始运行..

[2021-08-18 09:33:11.116415] INFO: moduleinvoker: 命中缓存

[2021-08-18 09:33:11.117622] INFO: moduleinvoker: input_features.v1 运行完成[0.006652s].

[2021-08-18 09:33:11.128787] INFO: moduleinvoker: dl_convert_to_bin.v2 开始运行..

[2021-08-18 09:33:11.133864] INFO: moduleinvoker: 命中缓存

[2021-08-18 09:33:11.135103] INFO: moduleinvoker: dl_convert_to_bin.v2 运行完成[0.006321s].

[2021-08-18 09:33:11.142224] INFO: moduleinvoker: dl_convert_to_bin.v2 开始运行..

[2021-08-18 09:33:11.148174] INFO: moduleinvoker: 命中缓存

[2021-08-18 09:33:11.149401] INFO: moduleinvoker: dl_convert_to_bin.v2 运行完成[0.00718s].

[2021-08-18 09:33:11.151351] INFO: moduleinvoker: dl_model_train.v1 开始运行..

[2021-08-18 09:33:11.156444] INFO: moduleinvoker: 命中缓存

[2021-08-18 09:33:11.157655] INFO: moduleinvoker: dl_model_train.v1 运行完成[0.006308s].

[2021-08-18 09:33:11.159593] INFO: moduleinvoker: dl_model_predict.v1 开始运行..

[2021-08-18 09:33:11.165187] INFO: moduleinvoker: 命中缓存

[2021-08-18 09:33:11.166459] INFO: moduleinvoker: dl_model_predict.v1 运行完成[0.00687s].

[2021-08-18 09:33:11.170667] INFO: moduleinvoker: cached.v3 开始运行..

[2021-08-18 09:33:11.176349] INFO: moduleinvoker: 命中缓存

[2021-08-18 09:33:11.177543] INFO: moduleinvoker: cached.v3 运行完成[0.006877s].

[2021-08-18 09:33:11.179660] INFO: moduleinvoker: instruments.v2 开始运行..

[2021-08-18 09:33:11.185654] INFO: moduleinvoker: 命中缓存

[2021-08-18 09:33:11.186971] INFO: moduleinvoker: instruments.v2 运行完成[0.007306s].

[2021-08-18 09:33:11.199996] INFO: moduleinvoker: hfbacktest.v1 开始运行..

[2021-08-18 09:33:11.204285] INFO: hfbacktest: biglearning V1.2.5

[2021-08-18 09:33:11.205427] INFO: hfbacktest: bigtrader v1.7.7

[2021-08-18 09:33:11.245812] INFO: moduleinvoker: cached.v2 开始运行..

[2021-08-18 09:33:11.253185] INFO: moduleinvoker: 命中缓存

[2021-08-18 09:33:11.254719] INFO: moduleinvoker: cached.v2 运行完成[0.008912s].

[2021-08-18 09:33:11.299679] INFO: moduleinvoker: cached.v2 开始运行..

[2021-08-18 09:33:11.314000] INFO: moduleinvoker: 命中缓存

[2021-08-18 09:33:11.315519] INFO: moduleinvoker: cached.v2 运行完成[0.015845s].

[2021-08-18 09:33:11.356146] INFO: moduleinvoker: cached.v2 开始运行..

[2021-08-18 09:33:11.369220] INFO: moduleinvoker: 命中缓存

[2021-08-18 09:33:11.370742] INFO: moduleinvoker: cached.v2 运行完成[0.014599s].

[2021-08-18 09:33:16.758998] INFO: hfbacktest: backtest done, raw_perf_ds:DataSource(d089616a4aed42268046552785b6b912T)

[2021-08-18 09:33:17.211878] INFO: moduleinvoker: hfbacktest.v1 运行完成[6.011888s].

[2021-08-18 09:33:17.213432] INFO: moduleinvoker: hftrade.v1 运行完成[6.024013s].

DataSource(ae2ab2082c654d19a51ae9eef0fdd97cT)

2021-08-18 09:33:11.438326 run trading v1.7.7 
2021-08-18 09:33:11.438452 init history datas... 
2021-08-18 09:33:11.457001 init trading env... 
in initial Future(4 [IF1906.CFX])
2019-06-03 09:31:00 开多！
2019-06-03 09:45:00 平多！
2019-06-03 09:50:00 开多！
2019-06-03 09:59:00 平多！
2019-06-03 10:08:00 开多！

收益率31.16%
年化收益率13089.38%
基准收益率5.56%
阿尔法0.35
贝塔5.27
夏普比率4.87
胜率0.2
盈亏比1.87
收益波动率111.73%
最大回撤11.41%

bigcharts-data-start/{"__type":"tabs","__id":"bigchart-b5d8ec808d42494bbbd6c15ef59c6080"}/bigcharts-data-end