行业轮动策略：
本策略每隔1个月定时触发计算1000能源（399381.SZ）、1000材料（399382.SZ）、1000工业（399383.SZ）、1000可选（399384.SZ）、1000消费（399385.SZ）、1000医药（399386.SZ）这几个行业指数过去20个交易日的收益率并选取了收益率最高的指数的成份股并获取了他们的市值数据随后把仓位调整至市值最大的5只股票上
该策略在股票指数日线下运行
获取股票池：
index_list = ['399381.SZ','399382.SZ','399383.SZ','399384.SZ','399385.SZ','399386.SZ']获取收益率最高的指数的成分股：
for k in list(his.keys()):
if len(his[k]) == 0:
del his[k]
for index in index_list:
ratio = 0
try:
ratio = (his[index][-2] - his[index][0])/his[index][0]
except KeyError:
print('key error:' + index)
except IndexError:
print('list index out of range:' + index)
return_index.append(ratio)
# 获取指定数内收益率表现最好的行业
best_index = index_list[np.argmax(return_index)]获取股票对应的市值：
# 获取当天有交易的股票
index_stock = ContextInfo.get_sector(best_index)
stock_available = []
for stock in index_stock:
if ContextInfo.is_suspended_stock(stock) == False:
stock_available.append(stock)
for stock in stock_available:
if stock in list(his.keys()):
#目前历史流通股本取不到，暂用总股本
if len(his[stock]) >= 2:
stocksize =his[stock][-2] * float(ContextInfo.get_financial_data(['CAPITALSTRUCTURE.total_capital'],[stock],lastdate,date).iloc[0,-1])
size_dict[stock] = stocksize
elif len(his[stock]) == 1:
stocksize =his[stock][-1] * float(ContextInfo.get_financial_data(['CAPITALSTRUCTURE.total_capital'],[stock],lastdate,date).iloc[0,-1])
size_dict[stock] = stocksize
else:
return
size_sorted = sorted(list(size_dict.items()), key = lambda item:item[1])
pre_holding = []策略代码：
#coding:gbk
'''
回测模型示例（非实盘交易策略）
本策略每隔1个月定时触发计算1000能源（399381.SZ）、1000材料（399382.SZ）、1000工业（399383.SZ）、
1000可选（399384.SZ）、1000消费（399385.SZ）、1000医药（399386.SZ）这几个行业指数过去
20个交易日的收益率并选取了收益率最高的指数的成份股并获取了他们的市值数据
随后把仓位调整至市值最大的5只股票上
该策略在股票指数日线下运行
'''
import numpy as np
import math
def init(ContextInfo):
MarketPosition ={}
ContextInfo.MarketPosition = MarketPosition #初始化持仓
index_universe = ['399381.SZ','399382.SZ','399383.SZ','399384.SZ','399385.SZ','399386.SZ']
index_stocks = []
for index in index_universe:
for stock in ContextInfo.get_sector(index):
index_stocks.append(stock)
ContextInfo.set_universe(index_universe+index_stocks) #设定股票池
ContextInfo.day = 20
ContextInfo.ratio = 0.8
ContextInfo.holding_amount = 5
ContextInfo.accountID='testS'
def handlebar(ContextInfo):
buy_condition = False
sell_condition = False
d = ContextInfo.barpos
lastdate = timetag_to_datetime(ContextInfo.get_bar_timetag(d - 1), '%Y%m%d')
date = timetag_to_datetime(ContextInfo.get_bar_timetag(d), '%Y%m%d')
print(date)
index_list = ['399381.SZ','399382.SZ','399383.SZ','399384.SZ','399385.SZ','399386.SZ']
return_index = []
weight = ContextInfo.ratio/ContextInfo.holding_amount
size_dict = {}
if (float(date[-4:-2]) != float(lastdate[-4:-2])):
#print '---------------------------------------------------------------------------------'
#print '当前交易日',date,date[-4:-2]
his = ContextInfo.get_history_data(21,'1d','close')
#print "his",his,timetag_to_datetime(ContextInfo.get_bar_timetag(d),"%Y%m%d")
for k in list(his.keys()):
if len(his[k]) == 0:
del his[k]
for index in index_list:
ratio = 0
try:
ratio = (his[index][-2] - his[index][0])/his[index][0]
except KeyError:
print('key error:' + index)
except IndexError:
print('list index out of range:' + index)
return_index.append(ratio)
# 获取指定数内收益率表现最好的行业
best_index = index_list[np.argmax(return_index)]
#print '当前最佳行业是：', ContextInfo.get_stock_name(best_index)[3:]+'行业'
# 获取当天有交易的股票
index_stock = ContextInfo.get_sector(best_index)
stock_available = []
for stock in index_stock:
if ContextInfo.is_suspended_stock(stock) == False:
stock_available.append(stock)
for stock in stock_available:
if stock in list(his.keys()):
#目前历史流通股本取不到，暂用总股本
if len(his[stock]) >= 2:
stocksize =his[stock][-2] * float(ContextInfo.get_financial_data(['CAPITALSTRUCTURE.total_capital'],[stock],lastdate,date).iloc[0,-1])
size_dict[stock] = stocksize
elif len(his[stock]) == 1:
stocksize =his[stock][-1] * float(ContextInfo.get_financial_data(['CAPITALSTRUCTURE.total_capital'],[stock],lastdate,date).iloc[0,-1])
size_dict[stock] = stocksize
else:
return
size_sorted = sorted(list(size_dict.items()), key = lambda item:item[1])
pre_holding = []
for tuple in size_sorted[-ContextInfo.holding_amount:]:
pre_holding.append(tuple[0])
#print '买入备选',pre_holding
#函数下单
if len(pre_holding) > 0:
sellshort_list = []
for stock in list(ContextInfo.MarketPosition.keys()):
if stock not in pre_holding and (stock in list(his.keys())):
order_shares(stock,-ContextInfo.MarketPosition[stock],'lastest',his[stock][-1],ContextInfo,ContextInfo.accountID)
print('sell',stock)
sell_condition = True
sellshort_list.append(stock)
if len(sellshort_list) >0:
for stock in sellshort_list:
del ContextInfo.MarketPosition[stock]
for stock in pre_holding:
if stock not in list(ContextInfo.MarketPosition.keys()):
Lots = math.floor(ContextInfo.ratio * (1.0/len(pre_holding)) * ContextInfo.capital / (his[stock][-1] * 100))
order_shares(stock,Lots *100,'lastest',his[stock][-1],ContextInfo,ContextInfo.accountID)
print('buy',stock)
buy_condition = True
ContextInfo.MarketPosition[stock] = Lots *100
#ContextInfo.paint('do_buy', int(buy_condition), -1, 0)
#ContextInfo.paint('do_sell', int(sell_condition), -1, 0)
