PyalgoTrade 優化（六）

時間  2019-12-11
標籤 pyalgotrade 優化 简体版
原文   原文鏈接

知足優化器組件。這個想法很簡單：python

有一個服務器負責：

  • 提供數據來運行策略。數組

  • 提供運行策略的參數。服務器

  • 記錄每一個工做線程的策略結果。函數

有多名工做人員負責：

  • 使用服務器提供的數據和參數運行策略。

爲了說明這一點，咱們將使用一種稱爲相對強弱指標RSI2的策略，
它須要如下參數：測試

  • SMA期間用於趨勢識別。咱們稱這個entrySMA爲150到250。
  • 退出點的SMA週期較小。咱們稱這個exitSMA爲5到15之間。
  • 進入短時間/長倉的RSI期間。咱們稱之爲rsiPeriod，範圍介於2到10之間。
  • 長期進倉的RSI超賣閾值。咱們稱此overSoldThreshold爲5到25之間。
  • RSI超買買入門檻。咱們稱之爲OverBoughtThreshold，範圍爲75到95。

若是個人數學是好的，那些是4409559不一樣的組合。優化

測試這個策略爲一組參數花了大約0.16秒。若是我連續執行全部組合，我須要大約8.5天的時間來評估全部組合，並找到最佳參數。那是很長一段時間，可是若是我可以拿到10臺8核電腦來完成這項工做，總時間將會降低到大約2.5個小時。
長話短說，咱們須要平行
咱們先從「道瓊斯工業平均水平」下載3年的每日k線數據：線程

python -c "from pyalgotrade.tools import yahoofinance; yahoofinance.download_daily_bars('dia', 2009, 'dia-2009.csv')"
python -c "from pyalgotrade.tools import yahoofinance; yahoofinance.download_daily_bars('dia', 2010, 'dia-2010.csv')"
python -c "from pyalgotrade.tools import yahoofinance; yahoofinance.download_daily_bars('dia', 2011, 'dia-2011.csv')"
from pyalgotrade import strategy
from pyalgotrade.technical import ma
from pyalgotrade.technical import rsi
from pyalgotrade.technical import cross


class RSI2(strategy.BacktestingStrategy):
    def __init__(self, feed, instrument, entrySMA, exitSMA, rsiPeriod, overBoughtThreshold, overSoldThreshold):
        super(RSI2, self).__init__(feed)
        self.__instrument = instrument
        # We'll use adjusted close values, if available, instead of regular close values.
        if feed.barsHaveAdjClose():
            self.setUseAdjustedValues(True)
        self.__priceDS = feed[instrument].getPriceDataSeries()
        self.__entrySMA = ma.SMA(self.__priceDS, entrySMA)
        self.__exitSMA = ma.SMA(self.__priceDS, exitSMA)
        self.__rsi = rsi.RSI(self.__priceDS, rsiPeriod)
        self.__overBoughtThreshold = overBoughtThreshold
        self.__overSoldThreshold = overSoldThreshold
        self.__longPos = None
        self.__shortPos = None

    def getEntrySMA(self):
        return self.__entrySMA

    def getExitSMA(self):
        return self.__exitSMA

    def getRSI(self):
        return self.__rsi

    def onEnterCanceled(self, position):
        if self.__longPos == position:
            self.__longPos = None
        elif self.__shortPos == position:
            self.__shortPos = None
        else:
            assert(False)

    def onExitOk(self, position):
        if self.__longPos == position:
            self.__longPos = None
        elif self.__shortPos == position:
            self.__shortPos = None
        else:
            assert(False)

    def onExitCanceled(self, position):
        # If the exit was canceled, re-submit it.
        position.exitMarket()

    def onBars(self, bars):
        # Wait for enough bars to be available to calculate SMA and RSI.
        if self.__exitSMA[-1] is None or self.__entrySMA[-1] is None or self.__rsi[-1] is None:
            return

        bar = bars[self.__instrument]
        if self.__longPos is not None:
            if self.exitLongSignal():
                self.__longPos.exitMarket()
        elif self.__shortPos is not None:
            if self.exitShortSignal():
                self.__shortPos.exitMarket()
        else:
            if self.enterLongSignal(bar):
                shares = int(self.getBroker().getCash() * 0.9 / bars[self.__instrument].getPrice())
                self.__longPos = self.enterLong(self.__instrument, shares, True)
            elif self.enterShortSignal(bar):
                shares = int(self.getBroker().getCash() * 0.9 / bars[self.__instrument].getPrice())
                self.__shortPos = self.enterShort(self.__instrument, shares, True)

    def enterLongSignal(self, bar):
        return bar.getPrice() > self.__entrySMA[-1] and self.__rsi[-1] <= self.__overSoldThreshold

    def exitLongSignal(self):
        return cross.cross_above(self.__priceDS, self.__exitSMA) and not self.__longPos.exitActive()

    def enterShortSignal(self, bar):
        return bar.getPrice() < self.__entrySMA[-1] and self.__rsi[-1] >= self.__overBoughtThreshold

    def exitShortSignal(self):
        return cross.cross_below(self.__priceDS, self.__exitSMA) and not self.__shortPos.exitActive()

服務器腳本code

import itertools
from pyalgotrade.barfeed import yahoofeed
from pyalgotrade.optimizer import server


def parameters_generator():
    instrument = ["dia"]
    entrySMA = range(150, 251)
    exitSMA = range(5, 16)
    rsiPeriod = range(2, 11)
    overBoughtThreshold = range(75, 96)
    overSoldThreshold = range(5, 26)
    return itertools.product(instrument, entrySMA, exitSMA, rsiPeriod, overBoughtThreshold, overSoldThreshold)

# The if __name__ == '__main__' part is necessary if running on Windows.
if __name__ == '__main__':
    # Load the feed from the CSV files.
    feed = yahoofeed.Feed()
    feed.addBarsFromCSV("dia", "dia-2009.csv")
    feed.addBarsFromCSV("dia", "dia-2010.csv")
    feed.addBarsFromCSV("dia", "dia-2011.csv")

    # Run the server.
    server.serve(feed, parameters_generator(), "localhost", 5000)

服務器代碼正在作3件事情：server

  • 聲明生成函數，爲該策略產生不一樣的參數組合。
  • 使用咱們下載的CSV文件加載Feed。
  • 運行端口5000上等待傳入鏈接的服務器。
    這是使用pyalgotrade.optimizer.worker模塊與服務器提供的數據並行運行策略的工做腳本：
from pyalgotrade.optimizer import worker
import rsi2

# The if __name__ == '__main__' part is necessary if running on Windows.
if __name__ == '__main__':
    worker.run(rsi2.RSI2, "localhost", 5000, workerName="localworker")

當您運行服務器和客戶端時，您將在服務器控制檯上看到相似的內容：get

2014-05-03 15:04:01,083 server [INFO] Loading bars
2014-05-03 15:04:01,348 server [INFO] Waiting for workers
2014-05-03 15:04:58,277 server [INFO] Partial result 1242173.28754 with parameters: ('dia', 150, 5, 2, 91, 19) from localworker
2014-05-03 15:04:58,566 server [INFO] Partial result 1203266.33502 with parameters: ('dia', 150, 5, 2, 81, 19) from localworker
2014-05-03 15:05:50,965 server [INFO] Partial result 1220763.1579 with parameters: ('dia', 150, 5, 3, 83, 24) from localworker
2014-05-03 15:05:51,325 server [INFO] Partial result 1221627.50793 with parameters: ('dia', 150, 5, 3, 80, 24) from localworker
.
.

在工做臺的控制檯上有這樣的東西：

2014-05-03 15:02:25,360 localworker [INFO] Running strategy with parameters ('dia', 150, 5, 2, 84, 15)
2014-05-03 15:02:25,377 localworker [INFO] Running strategy with parameters ('dia', 150, 5, 2, 94, 5)
2014-05-03 15:02:25,661 localworker [INFO] Result 1090481.06342
2014-05-03 15:02:25,661 localworker [INFO] Result 1031470.23717
2014-05-03 15:02:25,662 localworker [INFO] Running strategy with parameters ('dia', 150, 5, 2, 93, 25)
2014-05-03 15:02:25,665 localworker [INFO] Running strategy with parameters ('dia', 150, 5, 2, 84, 14)
2014-05-03 15:02:25,995 localworker [INFO] Result 1135558.55667
2014-05-03 15:02:25,996 localworker [INFO] Running strategy with parameters ('dia', 150, 5, 2, 93, 24)
2014-05-03 15:02:26,006 localworker [INFO] Result 1083987.18174
2014-05-03 15:02:26,007 localworker [INFO] Running strategy with parameters ('dia', 150, 5, 2, 84, 13)
2014-05-03 15:02:26,256 localworker [INFO] Result 1093736.17175
2014-05-03 15:02:26,257 localworker [INFO] Running strategy with parameters ('dia', 150, 5, 2, 84, 12)
2014-05-03 15:02:26,280 localworker [INFO] Result 1135558.55667
.
.

請注意，您應該只運行一個服務器和一個或多個工做。
若是您只想在本身的桌面上並行運行策略，您能夠利用pyalgotrade.optimizer.local 模塊，以下所示：

import itertools
from pyalgotrade.optimizer import local
from pyalgotrade.barfeed import yahoofeed
import rsi2


def parameters_generator():
    instrument = ["dia"]
    entrySMA = range(150, 251)
    exitSMA = range(5, 16)
    rsiPeriod = range(2, 11)
    overBoughtThreshold = range(75, 96)
    overSoldThreshold = range(5, 26)
    return itertools.product(instrument, entrySMA, exitSMA, rsiPeriod, overBoughtThreshold, overSoldThreshold)


# The if __name__ == '__main__' part is necessary if running on Windows.
if __name__ == '__main__':
    # Load the feed from the CSV files.
    feed = yahoofeed.Feed()
    feed.addBarsFromCSV("dia", "dia-2009.csv")
    feed.addBarsFromCSV("dia", "dia-2010.csv")
    feed.addBarsFromCSV("dia", "dia-2011.csv")

    local.run(rsi2.RSI2, feed, parameters_generator())

代碼正在作3件事情：
1.聲明生成不一樣參數組合的生成函數。
2.使用咱們下載的CSV文件加載Feed。
3.使用pyalgotrade.optimizer.local模塊並行運行策略，找到最佳結果。
當您運行此代碼時，您應該看到以下：

2014-05-03 15:08:06,587 server [INFO] Loading bars
2014-05-03 15:08:06,910 server [INFO] Waiting for workers
2014-05-03 15:08:58,347 server [INFO] Partial result 1242173.28754 with parameters: ('dia', 150, 5, 2, 91, 19) from worker-95583
2014-05-03 15:08:58,967 server [INFO] Partial result 1203266.33502 with parameters: ('dia', 150, 5, 2, 81, 19) from worker-95584
2014-05-03 15:09:52,097 server [INFO] Partial result 1220763.1579 with parameters: ('dia', 150, 5, 3, 83, 24) from worker-95584
2014-05-03 15:09:52,921 server [INFO] Partial result 1221627.50793 with parameters: ('dia', 150, 5, 3, 80, 24) from worker-95583
2014-05-03 15:10:40,826 server [INFO] Partial result 1142162.23912 with parameters: ('dia', 150, 5, 4, 76, 17) from worker-95584
2014-05-03 15:10:41,318 server [INFO] Partial result 1107487.03214 with parameters: ('dia', 150, 5, 4, 83, 17) from worker-95583
.
.

爲了記錄，發現的最佳結果是$ 2314.40，具備如下參數：

  • entrySMA：154
  • exitSMA：5
  • rsiPeriod：2
  • overBoughtThreshold：91
  • overSoldThreshold：18

做者：readilen連接：http://www.jianshu.com/p/8c43f54cf7a1來源：簡書著做權歸做者全部。商業轉載請聯繫做者得到受權，非商業轉載請註明出處。

相關文章
相關標籤/搜索
每日一句
    每一个你不满意的现在，都有一个你没有努力的曾经。
本站公眾號
   歡迎關注本站公眾號,獲取更多信息
聯繫我們 最近搜索 最新文章 沪ICP备13005482号-10 MyBatis教程 SQL 教程 MySQL教程 Java 教程 Thymeleaf 教程 Hibernate教程 Spring教程 Redis教程