同花顺Supermind量化交易 资金面专题2-风格下资金流入强度的分析 附源代码

一般而言，当市场中资金流动强度平缓时，风格投资者保持原有的风格方向；当资金流动强度强烈时，风格投资者转换原有的风格方向，例如：小市值投资者在资金流动强烈时，会转向大市值风格。本篇文章将从规模因子（大小盘风格）入手，研究关于资金流动强度与风格投资的话题。

风格下资金流入强度的分析

1 导入库包

In [2]:

import pandas as pd
import numpy as np
import datetime
import math
import matplotlib.pyplot as plt
plt.style.use('seaborn')

2.资金流入净值取数和计算

In [3]:

#取数平安银行主动买入量
data_buy=get_money_flow_step(security_list='000001.SZ',start_date=None,end_date='20190103',fre_step='1d',\
    fields=['act_buy_xl','act_buy_l','act_buy_m'],count=250,is_panel=0)['000001.SZ']

#取数平安银行主动卖出量
data_sell=get_money_flow_step(security_list='000001.SZ',start_date=None,end_date='20190103',fre_step='1d',\
    fields=['act_sell_xl','act_sell_l','act_sell_m'],count=250,is_panel=0)['000001.SZ']

#横向加总
data_buyall=pd.DataFrame(data_buy.apply(lambda x:x.sum(),axis=1),columns=['buy_all'])
data_sellall=pd.DataFrame(data_sell.apply(lambda x:x.sum(),axis=1),columns=['sell_all'])
data_all=pd.concat([data_buyall,data_sellall],axis=1)

#得出平安银行流入净值
data_all['data_net']=data_all['buy_all']-data_all['sell_all']
data_all.tail()

Out[3]:

	buy_all	sell_all	data_net
2018-12-26	108688097.0	176514956.0	-67826859.0
2018-12-27	191021237.0	258350579.0	-67329342.0
2018-12-28	249609737.0	161550282.0	88059455.0
2019-01-02	115789628.0	270478617.0	-154688989.0
2019-01-03	173588438.0	115469752.0	58118686.0

3.一个截面的资金流动强度分析

In [43]:

start_date='2019-01-10'
start_date=datetime.datetime.strptime(start_date,'%Y-%m-%d')

def inflow(stock_list,start_date):
    
    for stock in stock_list:
        #print(stock)
        data_buy=get_money_flow_step(security_list=stock,start_date=None,end_date=start_date,fre_step='1d',\
        fields=['act_buy_xl','act_buy_l','act_buy_m'],count=1,is_panel=0)[stock]
 
        data_sell=get_money_flow_step(security_list=stock,start_date=None,end_date=start_date,fre_step='1d',\
        fields=['act_sell_xl','act_sell_l','act_sell_m'],count=1,is_panel=0)[stock]
        
        data_buyall=pd.DataFrame(data_buy.apply(lambda x:x.sum(),axis=1),columns=['buy_all'])
        data_sellall=pd.DataFrame(data_sell.apply(lambda x:x.sum(),axis=1),columns=['sell_all'])
        data_all=pd.concat([data_buyall,data_sellall],axis=1) 
        data_net=data_all['buy_all']-data_all['sell_all']
        try:
            data_net=pd.Series(data_net[0],index=[stock])
        except:
            continue
        
        if 'inflow_df' in locals():
            inflow_df = inflow_df.append(data_net)
        else:
            inflow_df = data_net
    inflow_df=inflow_df.to_frame()
    inflow_df.columns=['net_inflow']

    return inflow_df
             
stock_list=list(get_all_securities('stock',date=start_date).index)
    
#取全市场股票各股票市值因子
df = get_fundamentals(query(asharevalue.symbol, asharevalue.total_mv).filter(\
                        asharevalue.symbol.in_(stock_list)).order_by(asharevalue.total_mv.asc()),date=start_date)
df.set_index('asharevalue_symbol',inplace=True)
df.index.name='symbol'
    
#取全市场股票各股票流入净值
df_inflow=inflow(stock_list,start_date)
    
#取全市场股票各股票成交量
df_volume = get_price(stock_list, None, '20170303', '1d', ['volume'], True, None, 1, is_panel=1)['volume']
df_volume=df_volume.T
df_volume.columns=['df_volume']
    
#合并三表
df=pd.concat([df,df_inflow,df_volume],axis=1)
    
#按照市值排序
df.sort_values(by='asharevalue_total_mv',ascending = True,inplace=True)
df.dropna(inplace=True)
df_slice=round(len(df)/10)
#print(df)
    
#将dataframe的净流入和成交量分成10份切片，并在切片内进行加总
data_net_inflow=[]
data_volume=[]
data_factor=[]
for i in range(10):
    data_sum=df.ix[i*df_slice:(i+1)*df_slice,['asharevalue_total_mv','net_inflow','df_volume']].apply(lambda x: x.sum())
    data_net_inflow.append(data_sum['net_inflow'])
    data_volume.append(data_sum['df_volume'])
    data_factor.append(data_sum['asharevalue_total_mv'])
    
data_net_inflow=pd.Series(data_net_inflow)
data_volume=pd.Series(data_volume)
data_factor=pd.Series(data_factor)
    
data_total=pd.DataFrame([data_factor,data_net_inflow,data_volume])
data_total=data_total.T
data_total.columns=['data_factor','data_net_inflow','data_volume']
data_total.head(10) 

Out[43]:

	data_factor	data_net_inflow	data_volume
0	4.776260e+11	-4.869376e+08	1.257051e+09
1	6.446934e+11	-1.012772e+09	1.832052e+09
2	8.006962e+11	-7.643456e+08	2.153484e+09
3	9.747939e+11	-9.009924e+08	2.541013e+09
4	1.179759e+12	-1.555130e+09	2.463412e+09
5	1.447312e+12	-1.324253e+09	2.500842e+09
6	1.