股票Screening- backtrader中文教程

发表于： 2019年9月18日 2019年12月25日
作者： hao chen
分类： backtrader, 量化交易与机器学习
标签： AAPL, Backtrader, cerebro, Concluding, IBM, itself, ORCL, Period, rets, Screener, Screening, script, self, SMA, st, TSLA, Usage, 原始, 回测系统

股票Screening

Looking一些其他的事情我碰到一个问题一提出的一个StackOverlow家族网站：定量Finance又名定量StackExchange。问题：

股票筛选和扫描开源软件使用的技术分析？

它被标记为Python，所以很有看头，如果backtrader是达task.

The分析itself

The问题似乎适用于一个简单的分析。虽然问题刚刚想要一上方均线，我们会继续像额外信息不符合条件的，个股以确保粮食实际上是被从chaff.

class Screener_SMA(bt.Analyzer):
    params = dict(period=10)

    def start(self):
        self.smas = {data: bt.indicators.SMA(data, period=self.p.period)
                     for data in self.datas}

    def stop(self):
        self.rets["over"] = list()
        self.rets["under"] = list()

        for data, sma in self.smas.items():
            node = data._name, data.close[0], sma[0]
            if data > sma:  # if data.close[0] > sma[0]
                self.rets["over"].append(node)
            else:
                self.rets["under"].append(node)

Note

Of当然，人们也需要分开import backtrader as bt

这几乎解决了这个问题。该Analyzer分析：

有period作为一个参数有一个灵活analyzer
start method
For每个data在系统做出Simple Moving Average（SMA），用于it.
stop method
Look其中data（close如果没有其他规定）高于其sma和存储在`关键list下的over回报（rets）

的构件rets是在标准analyzers和碰巧是collections.OrderedDict。创建由基class.

Keep不符合项下的标准under

现在的问题的那些：得到分析并running.

Note

我们假设代码已被放在一个名为st-screener.py

方法1

backtrader包括文件，因为时间已近开始，自动运行的脚本运行所谓的btrun，它可以装载战略，指标，从分析仪Python模块，分析论证，当然plot.

Let我们做一个运行：

$ btrun --format yahoo --data YHOO --data IBM --data NVDA --data TSLA --data ORCL --data AAPL --fromdate 2016-07-15 --todate 2016-08-13 --analyzer st-screener:Screener_SMA --cerebro runonce=0 --writer --nostdstats
===============================================================================
Cerebro:
  -----------------------------------------------------------------------------
  - Datas:
    +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
    - Data0:
      - Name: YHOO
      - Timeframe: Days
      - Compression: 1
    +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
    - Data1:
      - Name: IBM
      - Timeframe: Days
      - Compression: 1
    +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
    - Data2:
      - Name: NVDA
      - Timeframe: Days
      - Compression: 1
    +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
    - Data3:
      - Name: TSLA
      - Timeframe: Days
      - Compression: 1
    +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
    - Data4:
      - Name: ORCL
      - Timeframe: Days
      - Compression: 1
    +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
    - Data5:
      - Name: AAPL
      - Timeframe: Days
      - Compression: 1
  -----------------------------------------------------------------------------
  - Strategies:
    +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
    - Strategy:
      *************************************************************************
      - Params:
      *************************************************************************
      - Indicators:
        .......................................................................
        - SMA:
          - Lines: sma
          ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
          - Params:
            - period: 10
      *************************************************************************
      - Observers:
      *************************************************************************
      - Analyzers:
        .......................................................................
        - Value:
          - Begin: 10000.0
          - End: 10000.0
        .......................................................................
        - Screener_SMA:
          ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
          - Params:
            - period: 10
          ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
          - Analysis:
            - over: ("ORCL", 41.09, 41.032), ("IBM", 161.95, 161.221), ("YHOO", 42.94, 39.629000000000005), ("AAPL", 108.18, 106.926), ("NVDA", 63.04, 58.327)
            - under: ("TSLA", 224.91, 228.423)

我们已经使用了一组众所周知的行情：

AAPL, IBM, NVDA, ORCL, TSLA, YHOO

而这恰好是唯一一个是下10天Simple Moving
Average是TSLA.

让我们尝试了50天左右。是的，这也可以控制btrun。运行（输出缩短）：

$ btrun --format yahoo --data YHOO --data IBM --data NVDA --data TSLA --data ORCL --data AAPL --fromdate 2016-05-15 --todate 2016-08-13 --analyzer st-screener:Screener_SMA:period=50 --cerebro runonce=0 --writer --nostdstats
===============================================================================
Cerebro:
  -----------------------------------------------------------------------------
  - Datas:
    +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
    - Data0:
...
...
...
        - Screener_SMA:
          ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
          - Params:
            - period: 50
          ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
          - Analysis:
            - over: ("ORCL", 41.09, 40.339), ("IBM", 161.95, 155.0356), ("YHOO", 42.94, 37.9648), ("TSLA", 224.91, 220.4784), ("AAPL", 108.18, 98.9782), ("NVDA", 63.04, 51.4746)
            - under:

通知如何50天周期已在命令行中设定：

st-screener:Screener_SMA:period=50
在前面的运行，这是st-screener:Screener_SMA并且默认10从代码为used.

We也需要调整fromdate，以确保有足够的酒吧考虑了Simple Moving Averages

在这种情况下，所有代号都是above该50工作日移动average.

Approach 2

Craft一个小脚本（参见下面的完整代码）有更好的控制我们所做的。但结果是same.

其核心是相当小：

    cerebro = bt.Cerebro()
    for ticker in args.tickers.split(","):
        data = bt.feeds.YahooFinanceData(dataname=ticker,
                                         fromdate=fromdate, todate=todate)
        cerebro.adddata(data)

    cerebro.addanalyzer(Screener_SMA, period=args.period)
    cerebro.run(runonce=False, stdstats=False, writer=True)

约为参数解析其余mostly.

For10天（再次缩短输出）：

$ ./st-screener.py
===============================================================================
Cerebro:
...
...
...
        - Screener_SMA:
          ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
          - Params:
            - period: 10
          ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
          - Analysis:
            - over: (u"NVDA", 63.04, 58.327), (u"AAPL", 108.18, 106.926), (u"YHOO", 42.94, 39.629000000000005), (u"IBM", 161.95, 161.221), (u"ORCL", 41.09, 41.032)
            - under: (u"TSLA", 224.91, 228.423)

相同的结果。因此，让我们避免重蹈它50days.

Concluding

Both的btrun`从Approach 1和Approach 2使用小脚本一模一样的analyzer，因此提供相同results.

Andbacktrader已经能够承受另一个小challenge

Two最后说明：

这两种方法都使用内置的writer功能交付output.
- As参数btrun和--writer
- 作为参数传递给cerebro.run和writer=True
在这两种情况下runonce已停用。这是为了确保在线数据同步不断，因为结果可能有不同长度（股票的人能少买卖）

脚本usage

$ ./st-screener.py --help
usage: st-screener.py [-h] [--tickers TICKERS] [--period PERIOD]

SMA Stock Screener

optional arguments:
  -h, --help         show this help message and exit
  --tickers TICKERS  Yahoo Tickers to consider, COMMA separated (default:
                     YHOO,IBM,AAPL,TSLA,ORCL,NVDA)
  --period PERIOD    SMA period (default: 10)

The全script

#!/usr/bin/env python
# -*- coding: utf-8; py-indent-offset:4 -*-
###############################################################################
#
# Copyright (C) 2015, 2016 Daniel Rodriguez
#
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program.  If not, see <http://www.gnu.org/licenses/>.
#
###############################################################################
from __future__ import (absolute_import, division, print_function,
                        unicode_literals)

import argparse
import datetime

import backtrader as bt


class Screener_SMA(bt.Analyzer):
    params = dict(period=10)

    def start(self):
        self.smas = {data: bt.indicators.SMA(data, period=self.p.period)
                     for data in self.datas}

    def stop(self):
        self.rets["over"] = list()
        self.rets["under"] = list()

        for data, sma in self.smas.items():
            node = data._name, data.close[0], sma[0]
            if data > sma:  # if data.close[0] > sma[0]
                self.rets["over"].append(node)
            else:
                self.rets["under"].append(node)


DEFAULTTICKERS = ["YHOO", "IBM", "AAPL", "TSLA", "ORCL", "NVDA"]


def run(args=None):
    args = parse_args(args)
    todate = datetime.date.today()
    # Get from date from period +X% for weekeends/bank/holidays: let"s double
    fromdate = todate - datetime.timedelta(days=args.period * 2)

    cerebro = bt.Cerebro()
    for ticker in args.tickers.split(","):
        data = bt.feeds.YahooFinanceData(dataname=ticker,
                                         fromdate=fromdate, todate=todate)
        cerebro.adddata(data)

    cerebro.addanalyzer(Screener_SMA, period=args.period)
    cerebro.run(runonce=False, stdstats=False, writer=True)


def parse_args(pargs=None):

    parser = argparse.ArgumentParser(
        formatter_class=argparse.ArgumentDefaultsHelpFormatter,
        description="SMA Stock Screener")

    parser.add_argument("--tickers", required=False, action="store",
                        default=",".join(DEFAULTTICKERS),
                        help="Yahoo Tickers to consider, COMMA separated")

    parser.add_argument("--period", required=False, action="store",
                        type=int, default=10,
                        help=("SMA period"))

    if pargs is not None:
        return parser.parse_args(pargs)

    return parser.parse_args()


if __name__ == "__main__":
    run()