历史的十一月板块涨幅

来源：https://uqer.io/community/share/563c8065f9f06c713ddfeb6d

大盘又开始疯长了，还记嘚去年十一月吗 ？

#获得行业信息
def GetEquIndustry(universe,field):    
    num = 100
    cnt_num = len(universe)/num
    if cnt_num > 0:
        df = pd.DataFrame({})
        for i in range(cnt_num) :
            sub_df = DataAPI.EquIndustryGet(secID=universe[i*num:(i+1)*num],field=field)
            df = pd.concat([df,sub_df])
        if (i+1)*num != len(universe):
            sub_df = DataAPI.EquIndustryGet(secID=universe[(i+1)*num:],field=field)
            df = pd.concat([df,sub_df])
    else:
        df = DataAPI.EquIndustryGet(secID=universe,field=field)
    return df

from CAL.PyCAL import *
import pandas as pd
cal = Calendar('China.SSE')
universe = DataAPI.EquGet(equTypeCD='A')['secID'].tolist()    #获得全A股的secID
id2nm = lambda x:x[0:6]
tk_list_A = map(id2nm,universe)    #获得全A股的ticker

Ind_info = GetEquIndustry(universe = universe ,field=['ticker','secShortName','industryName2'])    #获得个股的申万行业分类
Ind_info_gp = Ind_info.groupby('industryName2')#按照行业分组

Ind_tks_dic = {}    #获得每个行业包含的股票
for ind_nm,sub_info in Ind_info_gp:
    Ind_tks_dic[ind_nm] = sub_info.drop_duplicates('ticker')['ticker'].tolist()

from pandas import DataFrame,Series
from CAL.PyCAL import *
cal = Calendar('China.SSE')
field = ['ticker','secShortName','tradeDate','preClosePrice','closePrice','turnoverValue']
#时间轴(开始时间)
time = ['20141031', '20131031', '20121031', '20111031', '20101031']
#保存各个时间段的数据
Data_time = {}
#保存各个时间段的股票名字
tk_nm_dic ={}

# 时间稍慢
for s in time : 
    Data_time[s] = DataFrame()
    for x in universe :        
        try : 
            data_temp = DataAPI.MktEqudAdjGet( secID = x , field =field , beginDate = s , endDate = cal.advanceDate(s,'1M', BizDayConvention.Following).strftime('%Y%m%d'))   
            data_temp['marketValue']  = DataAPI.MktEqudGet(secID = x ,field ='marketValue' , beginDate = s , endDate = cal.advanceDate(s,'1M', BizDayConvention.Following).strftime('%Y%m%d')) 
            Data_time[s] = pd.concat([Data_time[s],data_temp])  
        except : 
            continue
    tk_nm_dic[s] = dict(zip(Data_time[s]['ticker'],Data_time[s]['secShortName']))                                    # 获得个股ticker与名称的对应字典

for s in Data_time.values() :         
    s['tradeDate'] = pd.to_datetime(s['tradeDate'])                                         # 将tradeDate这一列的格式由string改为datetime
    s['increase'] = s['closePrice']/s['preClosePrice']                                      # 获得个股每天的收益

# 股票数据统计
Stock_Data = {}
for s in Data_time.keys() :    
    Stock_dict = {'ticker':[],'income':[],'turnoverValue':[] ,'marketValue' :[]} 
    # 获得每个时间段的Data计算个股的收益和平均市值
    for tk,sub_info in Data_time[s].groupby('ticker') :
        income = sub_info['increase'].prod()-1                     # 获得在这段时间内该股的涨幅    
        mkt_value = sub_info['marketValue'].sum()/len(sub_info)     
        turnoverValue_avg = sub_info['turnoverValue'].sum()/len(sub_info)   
        Stock_dict['ticker'].append(tk)
        Stock_dict['income'].append(income)
        Stock_dict['marketValue'].append(mkt_value)
        Stock_dict['turnoverValue'].append(turnoverValue_avg)
    # 返回时间为Key的个股数据
    Stock_Data[s] = pd.DataFrame(Stock_dict)

# 行业数据统计
Output_dicy = {}
Output_dicy['industry'] = []
Output_dicy['Num'] = []
Output_dicy['Nov14'] = []

for ind,tks in Ind_tks_dic.items() :
    for table in Stock_Data.keys() : 
        if not table in Output_dicy.keys() : 
            Output_dicy[table] = []
        sub_Industry = Stock_Data[table][Stock_Data[table]['ticker'].isin(tks)]      
         # 成交量前三
        bigstk = sub_Industry.sort(columns='turnoverValue',ascending=False)['ticker'][0:3].tolist()
        # 行业指数收益
        if not sub_Industry['marketValue'].sum() == 0 : 
            rtn_Industry = (sub_Industry['income']*sub_Industry['marketValue']).sum()/sub_Industry['marketValue'].sum()         
            Output_dicy[table].append(rtn_Industry)
            if table == '20141031' :   
                Output_dicy['Nov14'].append(map(lambda x:tk_nm_dic['20141031'][x],bigstk))

    if not sub_Industry['marketValue'].sum() == 0 :          
        #最新行业成分数量     
        Output_dicy['Num'].append(len(sub_Industry))
        Output_dicy['industry'].append(ind)       

Output_table = pd.DataFrame(Output_dicy)

去年十一月疯长的大盘

# 统计并显示
Out_put = Output_table.loc[:,['industry','Num','20101031','20111031','20121031','20131031','20141031','Nov14']]
Out_put.columns = [u'行业名称',u'该行业成分股数目(15年)',u'2010年十一月',u'2011年十一月',u'2012年十一月',u'2013年十一月',u'2014年十一月',u'2014年板块成交量前三']
Out_put[u'平均涨幅'] = (Out_put[u'2010年十一月']+Out_put[u'2011年十一月']+Out_put[u'2012年十一月']+Out_put[u'2013年十一月']+Out_put[u'2014年十一月']) / 5
print u'一共有%d个申万二级行业'%len(Out_put),u' ： '
Out_put.sort(u'2014年十一月' , ascending = False).head(20)

一共有208个申万二级行业  ：

五年间十一月平均涨幅

Out_put.sort(u'平均涨幅' , ascending = False).head(20)