In [246]:

import numpy as np
import pandas as pd

In [247]:

#获取股票代码列表
inst=D.instruments(start_date='2005-01-01', end_date=None, market='HK_STOCK')
len(inst)
inst[:5]

Out[247]:

['0001.HKEX', '0002.HKEX', '0003.HKEX', '0004.HKEX', '0005.HKEX']

In [248]:

#获取股票历史数据
df=pd.DataFrame({})
flag=True
for ins in inst:
    datas=D.history_data(instruments=ins, start_date='2014-01-01', end_date='2018-01-01',frequency='daily',fields= 'close')
    #print(ins)
    if type(datas)!=type(None) and not datas.empty and 'close' in datas.keys():
        df[ins]=datas['close']
        if flag:
            df['Date']=datas['date']
            flag=False
df.index=(df['Date'])
df.pop('Date')
df.head()

Out[248]:

	0001.HKEX	0002.HKEX	0003.HKEX	0004.HKEX	0005.HKEX	0006.HKEX	0007.HKEX	0008.HKEX	0009.HKEX	0010.HKEX	...	8480.HKEX	8481.HKEX	8485.HKEX	8491.HKEX	8495.HKEX	8559.HKEX	8587.HKEX	8590.HKEX	8591.HKEX	8593.HKEX
Date
2014-01-02	230.210571	263.047974	660.567871	149.237625	664.553894	214.540298	4.712500	22.975527	0.091450	106.948311	...	0.64	NaN	NaN	0.51	0.550	NaN	NaN	NaN	NaN	NaN
2014-01-03	225.898819	257.915344	646.481689	145.732040	651.592712	211.068771	4.793750	22.776318	0.091450	104.646896	...	0.67	NaN	0.32	0.51	0.530	NaN	NaN	0.077	NaN	NaN
2014-01-06	225.523880	257.273773	644.257568	143.353256	653.163757	207.944397	4.976562	22.709913	0.088500	103.970009	...	0.64	NaN	0.52	0.58	0.470	NaN	NaN	0.068	0.166058	6.3
2014-01-07	227.586029	256.846039	639.809326	143.854050	659.055176	205.167175	4.996875	22.975527	0.085550	103.699249	...	0.74	NaN	NaN	0.72	0.495	NaN	NaN	0.065	0.160769	6.3
2014-01-08	228.148422	258.556915	638.326538	146.107635	670.838074	206.035049	4.976562	23.506752	0.086533	104.511513	...	0.67	NaN	NaN	0.82	0.480	NaN	NaN	0.064	0.162885	6.3

5 rows × 2183 columns

In [249]:

#获取期间有空值的股票产品
row=df.shape[0]
empty=[]
for k in df.keys():
    tmp_df=pd.DataFrame(df[k].isnull().value_counts()).T
    if(False not in tmp_df.keys()):
        empty.append(k)
        continue
    tmp_row=tmp_df[False].values[0]
    if(tmp_row!=row):
        empty.append(k)

In [250]:

#删除有空值的产品
for k in empty:
    df.pop(k)
np.shape(df)

Out[250]:

(984, 1608)

In [251]:

#写入csv文件中
df.to_csv('HK_stock_Datas.csv')

In [252]:

#由于网站下载文件大小有限制，这里分成多个文件写入
for begin in np.arange(0,df.shape[1],100): 
    end=df.shape[1] if df.shape[1]<begin+100 else begin+100
    tmp_df=df.loc[:,df.keys()[begin:end]]
    tmp_df.to_csv('tmp'+str(int(begin/100))+'.csv')