up = stock_list[:100]
lower = stock_list[200:]
random.seed(50)
sup = random.sample(up, 20)
slow = random.sample(lower, 10)
r = sup + slow
print(r)
test_list = ['2888']
for name in test_list:
#for name in r:
s = query_part(name=name,
feature='close',
adjust='all',
collection_name='HKEX_ADJ',
db_name='MarketDataUpdate')
# s['strdate'] = pd.to_datetime(s['date'], unit='s')
s['strdate'] = pd.to_datetime(s['date'], unit='s')
s['strdate'] = s['strdate'].apply(lambda x: x.strftime('%Y-%m-%d'))
s['rebase_mongodb'] = s['value'] / s['value'][0]
yahoo = pd.read_csv(
'/home/pingan/PycharmProjects/test/calculate_AF/yahoo/' +
name.zfill(4) + '.HK.csv')
# yahoo = yahoo.drop(yahoo.index[0])
yahoo = yahoo[yahoo['Date'] >= '2018-07-09']
yahoo['rebase_yahoo'] = yahoo['Adj Close'] / yahoo['Adj Close'].tolist()[0]
# print(type(s['strdate'].tolist()[0]))
# print(type(yahoo['Date'].tolist()[0]))
path = '/home/pingan/PycharmProjects/test/calculate_AF/adj_price_without_adjustment_factor'
without = getfilename(path, '.csv')
print(without)
test_list = ['1451']
equal_count = []
not_equal_ticker = []
#notequal_df = pd.DataFrame()
#for name in test_list:
for name in without:
check_column = []
adj = query_part(name=name,
feature={'$in': ['close', 'high', 'open', 'low']},
adjust='all',
collection_name='HKEX_ALL',
db_name='MarketData')
noa = query_part(name=name,
feature={'$in': ['close', 'high', 'open', 'low']},
adjust='no',
collection_name='HKEX_ALL',
db_name='MarketData')
if adj.empty or noa.empty:
continue
else:
new = pd.merge(noa,
adj,
on=['date', 'feature', 'name', 'category'],
how='outer')
import pymongo
import sys
sys.path.append('/home/pingan/PycharmProjects/test/DataUpdate/tools')
from mongodb import query_all, query_part, tomongo
import pandas as pd
import os
import matplotlib.pyplot as plt
import pylab
import xlrd
old_factor_file = query_part(feature='price_af_star',
collection_name='HKEX_AF_Daily',
db_name='MarketDataUpdate')
adjustment_list = old_factor_file['name'].values.tolist()
# for name in ticker_list:
# new_factor = query_part(name = name, feature = 'price_af', collection_name = 'HKEX_AF_Daily', db_name = 'MarketDataUpdate')
# new_factor.to_csv('/home/pingan/PycharmProjects/test/calculate_AF/new_daily_factor/' + name + '.csv')
# no_adjust = query_part(name = name, feature = {'$in':['close', 'open', 'high', 'low']}, collection_name = 'HKEX_ALL_NOADJ', db_name = 'MarketDataUpdate')
# no_adjust.to_csv('/home/pingan/PycharmProjects/test/calculate_AF/no_adjust_price/' + name + '.csv')
# calculate ticker who has adjustmentfactor
adjustment_list = [x for x in adjustment_list]
print(adjustment_list)
test_list = ['1136']
for name in adjustment_list:
#for name in test_list:
import pymongo
import sys
sys.path.append('/home/pingan/PycharmProjects/test/DataUpdate/tools')
from mongodb import query_all, query_part
list = [
'8096', '4335', '4336', '6860', '667', '8017', '2168', '3868', '1832',
'1775', '1743', '1025'
]
for i in range(len(list)):
data = query_part(name=list[i],
feature='close',
adjust='no',
collection_name='HKEX_ALL',
db_name='MarketData')
data['date'] = pd.to_datetime(data['date'], unit='s')
data.to_csv(os.path.join(
'/home/pingan/PycharmProjects/test/hkex_noadj/abnormal_start_list/' +
list[i] + '.csv'),
index=False)
#667 HK Equity 2006-10-05-----------2011-12-08
#1025 2003-11-20----------2016-01-06
#1743 2007-07-06------------2008-06-16
#1775 2008-04-02---------2008-11-10
#1832 2007-07-13------------2010-05-24
#2168 this ticker is belongs to another company but it died and kasai own this ticker now, so the data start from 2018-12-05
#3868 2007-10-02-----------2017-11-29