谁是中国A股最有钱的自然人

来源：https://uqer.io/community/share/5523b45ef9f06c8f3390453e

运行此代码，便可知道，在当前日期，谁是A股最有钱的自然人！

股东数据来自于恒生聚源，选择的是类型为“自然人”的股东，有可能有脏数据！

import pandas as pd
import numpy as np
from datetime import datetime,timedelta
from CAL.PyCAL import *
cal = Calendar('China.SSE')

def GetSecID(tk_list,**kargs):    #获得partyID
    num = 100
    cnt_num = len(tk_list)/num
    if cnt_num > 0:
        df = pd.DataFrame({})
        for i in range(cnt_num):
            sub_df = DataAPI.SecIDGet(ticker=tk_list[i*num:(i+1)*num],**kargs)
            df = pd.concat([df,sub_df])
        if (i+1)*num != len(tk_list):
            sub_df = DataAPI.SecIDGet(ticker=tk_list[(i+1)*num:],**kargs)
            df = pd.concat([df,sub_df])
    else:
        df = DataAPI.SecIDGet(ticker=tk_list,**kargs)
    return df

def CountTime():    #获取最近的一个交易日，返回的是datetime格式    
    today = datetime.today()
    today_str = today.strftime("%Y%m%d")
    cal_date = Date.fromDateTime(today)       
    time1=" 15:05:00"
    ben_time = datetime.strptime(today_str+time1,"%Y%m%d %H:%M:%S")
    if cal.isBizDay(cal_date) & (today>ben_time):    #如果是交易日，则判断当天是不是在15点前        
        date = today
    else:    #如果当天不是交易日，则获得前一个交易日
        cal_wd = cal.advanceDate(cal_date, '-1B', BizDayConvention.Following)    #Date格式
        date = cal_wd.toDateTime()    #datetime格式        
    return date
def GetMktEqud(tk_list,**kargs):    #获得最新市场信息快照，即最新价格信息
    num = 50
    cnt_num = len(tk_list)/num
    if cnt_num > 0:
        df = pd.DataFrame({})
        for i in range(cnt_num):
            sub_df = DataAPI.MktEqudGet(ticker=tk_list[i*num:(i+1)*num],**kargs)
            df = pd.concat([df,sub_df])
        if (i+1)*num != len(tk_list):
            sub_df = DataAPI.MktEqudGet(ticker=tk_list[(i+1)*num:],**kargs)
            df = pd.concat([df,sub_df])
    else:
        df = DataAPI.MktEqudGet(ticker=tk_list,**kargs)
    return df

def add_nm_money(sub_info):    #将个股名称与金额拼接，方便做展示
    add_info_list = []
    sub_info_1 = sub_info.sort(columns='hold_money',ascending=False)
    for i in range(len(sub_info_1)):
        add_info = sub_info_1['secshortNM'].iloc[i] + str(round(sub_info_1['hold_money'].iloc[i]/1e8,2))+'亿'
        add_info_list.append(add_info)
    return add_info_list

#获得全A股的partyID

universe = DataAPI.EquGet(equTypeCD='A')['secID'].tolist()    #获得全A股的secID
All_A_tks_list = map(lambda x:x[0:6],universe)                #根据色此ID获得A股的所有ticker，因为要获得partyID需要输入ticker
party_id_info = GetSecID(tk_list=All_A_tks_list,field=['secShortName','ticker','partyID'])    #由ticker获得该个股的partyID
party2tk_dic = dict(zip(party_id_info['partyID'],party_id_info['ticker']))          #获得partyID与ticker的对应字典；注意，party_id_info的partyID是int型
party2nm_dic = dict(zip(party_id_info['partyID'],party_id_info['secShortName']))    #获得partyID与secShortName的对应字典
party_id_list = map(lambda x:str(x),party_id_info['partyID'].tolist())     #获得partyID的list，返回的party_id_info的‘partyID’是int型，而EquMainshJYGet输入的partyID需要str型，所以这里做个转换
field1 = ['partyID','publishDate','shName','shChar','holdVol']    #分别对应的是[公司代码，公告日、信息类别、股东名称、股东性质、持股数]

#获得所有个股的自然人股东姓名，以及持有的股票数目
All_info_df = pd.DataFrame({})
for party_id in party_id_list:
    hold_info = DataAPI.JY.EquMainshJYGet(partyID=party_id,field=field1)
    last_publishDate = hold_info['publishDate'].iloc[-1]
    hold_info = hold_info[(hold_info['publishDate']==last_publishDate)&(hold_info['shChar']=='自然人')]    #获取最新的自然人股东信息
    hold_info_gp = hold_info.groupby('shName')
    #由于EquMainshJYGet这个API获得的是十大股东和十大流通股，只要出现过，不管是哪种都要记录；有可能出现两次，也只记录一次。
    for nm,sub_info in hold_info_gp:    
        if len(sub_info)>1:   #既是十大股东之一也是十大流通股东之一，只记录其中之一         
            All_info_df = pd.concat([All_info_df,sub_info[0:1]])
        else:    #是十大股东或十大流通股东，记录下来
            All_info_df = pd.concat([All_info_df,sub_info])
All_info_df['ticker'] = All_info_df['partyID'].apply(lambda x:party2tk_dic[x])    #获得partyID对应的ticker

#获得个股的行情数据
tklist_1 = All_info_df['ticker'].tolist()    #获得有自然人持股的个股ticker
tklist_1 = list(set(tklist_1))    #去重
endDate = CountTime().strftime('%Y%m%d')    #获得最近一个交易日的日期
Mkt_info = GetMktEqud(tklist_1,beginDate=endDate,endDate=endDate,field = ['ticker','closePrice'])    #获取最近一个交易日的行情数据

tk2price = dict(zip(Mkt_info['ticker'],Mkt_info['closePrice']))    #获得个股ticker与价格的字典
All_info_df['closePrice'] = All_info_df['ticker'].apply(lambda x:tk2price[x])    #添加closePrice到All_info_df中
All_info_df['secshortNM'] = All_info_df['partyID'].apply(lambda x:party2nm_dic[x])    #添加secshortNM即个股的简称到All_info_df中
All_info_df['hold_money'] = All_info_df['holdVol']*All_info_df['closePrice']    #添加hold_money即个股的持有金额到All_info_df中
All_info_df_gp = All_info_df.groupby('shName')    #根据股东的姓名来分类

#统计自然人股东的总资产，并按照总资产大小由大到小排序
final_info_dic = {'name':[],'total_money':[],'stk_money':[]}
for personNM,sub_info in All_info_df_gp:
    total_money = sub_info['hold_money'].sum()
    stk_money = add_nm_money(sub_info)    #获得个股名称和金额的拼接结果    
    final_info_dic['name'].append(personNM)
    final_info_dic['total_money'].append(total_money)
    final_info_dic['stk_money'].append(stk_money)   
final_info_df = pd.DataFrame(final_info_dic)

#为了展示，做一些处理
All_info_df_sort = final_info_df.sort(columns='total_money',ascending=False).reset_index(drop=True)
All_info_df_sort['total_money'] = np.round(All_info_df_sort['total_money']/1e8,2).astype(str)+'亿'
All_info_df_sort.columns = ['自然人名称','持有的股票及资产','总资产']
print '谁是A股最有钱的自然人股东？（附注：’恒生聚源‘的数据库显示为自然人，则该股东定为自然人股东，可能存在脏数据）'
All_info_df_sort

谁是A股最有钱的自然人股东？（附注：’恒生聚源‘的数据库显示为自然人，则该股东定为自然人股东，可能存在脏数据）

11216 rows × 3 columns