原文策略源码如下：
#趋势跟踪策略

导入函数库

import jqdata
import talib
import pandas as pd
import numpy as np
import datetime
from datetime import timedelta
import math
from numpy import nan

初始化函数，设定基准等等

def initialize(context):
# 设定沪深300作为基准
set_benchmark(‘000001.XSHG’)
# 开启动态复权模式(真实价格)
set_option(‘use_real_price’, True)
# 输出内容到日志 log.info()
log.info(‘初始函数开始运行且全局只运行一次’)
# 过滤掉order系列API产生的比error级别低的log
# log.set_level(‘order’, ‘error’)
g.num_stock = 10
g.index = ‘000001.XSHG’
g.index2 = ‘000300.XSHG’#沪深300
g.index8 = ‘399006.XSHE’#创业板指数
# g.index2 = ‘000016.XSHG’#上证50
# g.index8 = ‘399333.XSHE’#中小板R
g.risk_ratio = 0.10
### 股票相关设定 ###
# 股票类每笔交易时的手续费是：买入时佣金万分之三，卖出时佣金万分之三加千分之一印花税, 每笔交易佣金最低扣5块钱
set_order_cost(OrderCost(close_tax=0.001, open_commission=0.0003, close_commission=0.0003, min_commission=5), type=‘stock’)
run_monthly(rebalence,1,time = ‘14:50’)

开盘前运行函数

def before_trading_start(context):

开盘时运行函数

def handle_data(context, data):
if market_not_safe(g.index2,g.index8):
# if len(context.portfolio.positions) >0:
clear_position(context)
return
# if g.days%g.period == 0:
# if [context.current_dt.hour,context.current_dt.minute] == [14,50]:
# rebalence(context)
# g.days = g.days+1
# for stock in context.portfolio.positions:
# if can_sell(stock):
# order_target_value(stock,0)

def rebalence(context):
# if not market_not_safe(g.index2,g.index8):

stocklist = select_stock(context)
#卖出
for stock in context.portfolio.positions:
    if stock not in stocklist:
        order_target_value(stock,0)
stock_values = calPosition(context,stocklist)
for stock in stock_values:
    curPrice = attribute_history(stock, 1, '1m', ('close'),fq='pre')['close'].values
    curValue = context.portfolio.positions[stock].total_amount * curPrice
    target_value = stock_values[stock]
    if abs(target_value - curValue)>0.2:
        order_target_value(stock,target_value)

def select_stock(context):
stocklist = get_index_stocks(g.index)
stock_list = filter_specials(stocklist)
stock_list = filter_new_and_sub_new(stock_list)
stock_list = filter_big_amout_of_increase(stock_list)
factor_dict = {}
for stock in stock_list:
temp = []
temp.append(get_smart_money_factor(stock))
factor_dict[stock] = temp
df_factor = pd.DataFrame(factor_dict).T
df_factor = df_factor.dropna()
df_factor.columns = [‘smart_money_factor’]
df = df_factor.sort(‘smart_money_factor’)
print df.head()
# stock_list = list(df.index.values)[:int(len(df)*0.1)]
stock_list = list(df.index.values)[:10]

return stock_list

def filter_big_amout_of_increase(stocklist,ratio = 0.2):
stock_dict = {}
for stock in stocklist:
stock_dict[stock] = get_growth_rate(stock)
sort_list = sorted(stock_dict.items(),key = lambda x:x[1])
stock_list = sort_list[:int(len(sort_list)*(1-ratio))]
stock_list = [stock[0] for stock in stock_list]
return stock_list

def filter_new_and_sub_new(stocklist,days = 60):
stock_list = []
for stock in stocklist:
start_date = get_security_info(stock).start_date
if (datetime.date.today()-start_date)>timedelta(60):
stock_list.append(stock)
return stock_list

def get_smart_money_factor(stock,count_day = 1):

df = attribute_history(stock,count_day*230,'1m',['open','close','volume'])
df['zdf'] = abs(df['close']/df['open']-1)
df['smart'] = df['zdf']/(df['volume'].apply(math.sqrt))
df = df.sort('smart',ascending = False)
df['cum_vol'] = df['volume'].cumsum()
df['cum_vol'] = df['cum_vol']/df.ix[-1,'cum_vol']
df_smart = df[df['cum_vol']<=0.2]
all_vol = df['volume'].sum()
all_smartvol = df_smart['volume'].sum()
VWAP_smart = (df_smart['volume']*df_smart['close']/all_smartvol).sum()
VWAP_all = (df['volume']*df['close']/all_vol).sum()
try:
    factor = VWAP_smart/VWAP_all
except:
    factor = nan
return factor

def filter_specials(stock_list):
curr_data = get_current_data()
stock_list = [stock for stock in stock_list if
(not curr_data[stock].paused) # 未停牌
and (not curr_data[stock].is_st) # 非ST
and (‘ST’ not in curr_data[stock].name)
and (’*’ not in curr_data[stock].name)
and (‘退’ not in curr_data[stock].name)
and (curr_data[stock].low_limit < curr_data[stock].day_open < curr_data[stock].high_limit) # 未涨跌停
]

return stock_list

def market_not_safe(index2,index8):
gr_index2 = get_growth_rate(index2)
gr_index8 = get_growth_rate(index8)
if gr_index2 <= 0 and gr_index8 <= 0:
return True

def get_growth_rate(security, n=20):
lc = get_close_price(security, n)
c = get_close_price(security, 1, ‘1m’)

#判断是数据是否为空
if not isnan(lc) and not isnan(c) and lc != 0:
    return (c - lc) / lc
else:
    log.error("数据非法, security: %s, %d日收盘价: %f, 当前价: %f" %(security, n, lc, c))
    return 0

def can_buy(security,short=5,long=20):
#检查均线是否多头
ma_short = attribute_history(security, short, ‘1d’, (‘close’), True).mean()[‘close’]
ma_long = attribute_history(security, long, ‘1d’, (‘close’), True).mean()[‘close’]
cur_price = get_close_price(security,1,‘1m’)
yes_price = get_close_price(security,1)
if ma_short >= ma_long and cur_price > yes_price:
return True

def get_close_price(security, n, unit=‘1d’):
return attribute_history(security, n, unit, (‘close’), True,fq=‘pre’)[‘close’][0]

def clear_position(context):
if len(context.portfolio.positions)>0:
for stock in context.portfolio.positions:
order_target_value(stock,0)
log.info(‘大盘止损,全盘清仓’)

def calPosition(context, buylist):
# 每次调仓，用 positionAdjustFactor 来控制承受的风险，最大损失为总仓位 * positionAdjustFactor
positionAdjustValue = context.portfolio.total_value * g.risk_ratio
Ajustvalue_per_stock = float(positionAdjustValue)/len(buylist)

hStocks = history(1, '1m', 'close', buylist, df=False)

risk_value = {}
# 计算组合后的 ATR，此处假设资金是平均分配。
for stock in buylist:
    curATR = 2*float(fun_getATR(stock))
    if curATR != 0 :
        risk_value[stock] = hStocks[stock]*Ajustvalue_per_stock/curATR
    else:
        risk_value[stock] = 0

total_value = sum([risk_value[stock] for stock in risk_value])
if total_value > context.portfolio.total_value:
    tempdict = {}
    for stock in buylist:
        tempdict[stock] = context.portfolio.total_value*(risk_value[stock]/total_value)
    risk_value = tempdict
return risk_value

def fun_getATR(stock):
ATRlag = 14

# 计算ATR及Keltner Band 
try:
    hStock = attribute_history(stock, ATRlag+10, '1d', ('close','high','low') , df=False)
except:
    log.info('%s 获取历史数据失败' %stock)
    return 0
close_ATR = fun_normalizeData(hStock['close'])
high_ATR = fun_normalizeData(hStock['high'])
low_ATR = fun_normalizeData(hStock['low'])
try:
    ATR = talib.ATR(high_ATR, low_ATR, close_ATR, timeperiod=ATRlag)
except:
    return 0

return ATR[-1]

def fun_normalizeData(myData):
# 去极值
std = np.std(myData)
MA = np.mean(myData)
for i in range(len(myData)):
if myData[i] > MA + 3std:
myData[i] = MA + 3std
elif myData[i] < MA - 3std:
myData[i] = MA - 3std
return myData

来源：CSDN

作者：程序化小学生

链接：https://blog.csdn.net/weixin_45131220/article/details/103489486

标签

源码

聚宽源码36

导入函数库

初始化函数，设定基准等等

开盘前运行函数

def before_trading_start(context):

开盘时运行函数