def search(txt: str, n: int):
"""Função que coleta as 'n' primeiras buscas referentes a
txt = 'tesouro' ou txt = 'bvsp'.
Args:
txt (str): objeto a ser pesquisado: 'tesouro' ou 'bvsp'.
n (int): número de resultados.
Returns:
iv..utils.search_obj.SearchObj: iterator de dicionários
"""
pdt = []
if txt == 'tesouro':
pdt = ['bonds']
elif txt == 'bvsp':
pdt = ['indices']
search_results = iv.search_quotes(
text=txt,
products=pdt,
countries=['brazil'],
n_results=n
)
return search_results
def technical(symbol):
ta = investpy.search_quotes(text=symbol,
products=['stocks'],
countries=['united states'],
n_results=1)
today = datetime.today().strftime('%d/%m/%Y')
t_indicator_1d = ta.retrieve_technical_indicators(interval="daily")
t_indicator_1d.to_csv("ta_1d.csv", index=False)
def test_investpy_search():
"""
This function checks that investpy search function works properly.
"""
params = [{
'text': 'bbva',
'products': None,
'countries': None,
'n_results': 5
}, {
'text': 'spain 3y',
'products': None,
'countries': None,
'n_results': 5
}, {
'text': 'ibex 35',
'products': None,
'countries': None,
'n_results': 5
}, {
'text': 'bnp daxplus',
'products': None,
'countries': None,
'n_results': None
}, {
'text': 'apple',
'products': ['stocks'],
'countries': ['united states'],
'n_results': 1
}]
for param in params:
results = investpy.search_quotes(text=param['text'],
products=param['products'],
countries=param['countries'],
n_results=param['n_results'])
dates = [
{
'from_date': '01/01/2018',
'to_date': '01/01/2019'
},
{
'from_date': '01/01/1990',
'to_date': '01/01/2019'
},
]
for result in results:
print(result)
result.retrieve_recent_data()
for date in dates:
result.retrieve_historical_data(from_date=date['from_date'],
to_date=date['to_date'])
break
def get_data(name, prod, start_date, end_date):
k1 = investpy.search_quotes(text=name,
products=[prod],
countries=['India'],
n_results=2)[0].retrieve_historical_data(
from_date=start_date, to_date=end_date)
stockfile = name + '_' + prod + '.pkl'
k1.to_pickle(stockfile)
return k1
print('stock data downloaded')
def get_stock_data(name):
k1 = investpy.search_quotes(text=name,
products=['stocks'],
countries=['India'],
n_results=2)[0]
k1 = investpy.get_stock_historical_data(stock=k1.symbol,
country='India',
from_date='01/01/2015',
to_date='20/03/2021')
return k1
def pesquisa(self, ativo, pais=None):
if pais is not None and not isinstance(pais, str):
return "ERR#001: país especificado inválido"
if ativo is None:
return "ERR#002: Ativo não pode ser vazio"
try:
dados = investpy.search_quotes(text=ativo, countries=[self.pais])
except ValueError:
return "ERR#003: Ativo " + ativo + " não encontrado"
values = {}
for dado in dados:
values = dado.__dict__
return values
def get_historical_values(self, from_date, to_date):
"""Function that gets the historical closing value of the asset at investing.com
through the investpy.
Args:
from_date (str): start date of the series in the format %d/%m/%Y
to_date (str): ending date of the series in the format %d/%m/%Y
Returns:
pd.Series: of the historical values from_date to_date
"""
engine = investpy.search_quotes(text=self.code,
products=["funds"],
countries=["brazil"])
closings = engine.retrieve_historical_data(from_date=from_date,
to_date=to_date)["Close"]
return closings
def update_stock_price_history(self, ticker, country):
# It today is not a business day, instead set use the latest b.day
today = dt.datetime.now()
today_m7 = (today - dt.timedelta(days=7)).strftime('%Y-%m-%d')
today = (pd.bdate_range(
start=today_m7,
end=today.strftime('%Y-%m-%d'))[-1].strftime('%d/%m/%Y'))
# By default get data starting 5 years ago
begin = (dt.datetime.strptime(today, '%d/%m/%Y') -
dt.timedelta(days=365 * 5)).strftime('%d/%m/%Y')
path = 'price_history\\{}.csv'.format(ticker)
# If some data was already gotten, start from the latest date
if os.path.isfile(path):
data0 = pd.read_csv(path, index_col=0, header=0, sep=';')
data0.index = pd.to_datetime(data0.index)
begin = (data0.index[-1] +
dt.timedelta(days=1)).strftime('%d/%m/%Y')
else:
data0 = pd.DataFrame(columns=['Price'],
index=pd.Series([], name='Date'))
if (dt.datetime.strptime(begin, '%d/%m/%Y') < dt.datetime.strptime(
today, '%d/%m/%Y')):
print('Updating history for {}...'.format(ticker))
# try:
# data = investpy.get_stock_historical_data(stock=ticker,
# country=country,
# from_date=begin,
# to_date=today)
# except:
try:
search_result = investpy.search_quotes(text=ticker,
countries=[country],
n_results=1)
data = search_result.retrieve_historical_data(from_date=begin,
to_date=today)
data.index = pd.to_datetime(data.index)
data = data[['Close']]
data = data.rename(columns={'Close': 'Price'})
data = pd.concat([data0, data], axis=0).sort_index()
data.index = data.index.strftime('%Y-%m-%d')
data = data.sort_index()
data.to_csv(path, index=True, header=True, sep=';')
except:
pass
def get_hist_prices(pf, days_back=0):
print('Retrieving prices...')
if days_back == 0:
from_date = pf['date'].min().strftime('%d/%m/%Y')
elif days_back > len(_prices):
print("More back than existing prices")
from_date = (exclude_weekends() - datetime.timedelta(days=days_back)).strftime('%d/%m/%Y')
else:
print("prices already exist")
return _prices.iloc[-days_back:]
# from_date = _prices.iloc[-1].Date.strftime('%d/%m/%Y')
to_date = exclude_weekends().date().strftime('%d/%m/%Y') #(exclude_weekends() - datetime.timedelta(days=1))
all_prices = []
pf = shrink_pf(pf)
for i, row in pf.iterrows():
search_results = investpy.search_quotes(text=row['ticker'], products=[row['product']], countries=[row['country']])
# from_date = row['date'].strftime('%d/%m/%Y')
for sr in search_results[:1]:
hist_p = sr.retrieve_historical_data(from_date=from_date, to_date=to_date)['Close'].to_frame().rename({'Close':row['ticker']}, axis=1)
all_prices.append(hist_p.interpolate(axis=1))
return pd.concat(all_prices, axis=1)
from scipy.special import gamma
from arch import arch_model
import pyflux as pf
import sys
import investpy
from sklearn.ensemble import AdaBoostClassifier, RandomForestClassifier
import feats
from supervised.automl import AutoML
#Main.include("try3.jl")
#pos=int(input("enter stock pos: "))
#lb = int(input("enter no of returns: "))
#k=pd.read_pickle('/home/sahil/projdir/dailydata.pkl')
#k1=k[k.Symbol==k.Symbol.unique()[pos]].iloc[-lb:]
automl = AutoML(mode='Compete', total_time_limit=600)
k1 = investpy.search_quotes(text='Kotak NIFTY ETF',
products=['etfs'],
countries=['India'],
n_results=2)[0]
k1 = investpy.get_etf_historical_data('KOTAKNIFTY',
country='India',
from_date='01/01/2010',
to_date='20/03/2021')
#k1=investpy.search_quotes(text='AARTIIND',products=['stocks'],countries=['India'],n_results=2)[0].retrieve_historical_data(from_date='01/01/2019',to_date='07/12/2020')
k2 = investpy.get_index_historical_data(index='India VIX',
country='India',
from_date='01/01/2010',
to_date='20/03/2021')
fund='bbva plan multiactivo moderado pp',
country='spain',
from_date='01/01/2010',
to_date='01/01/2019')
df_bbva_etf = investpy.get_etf_recent_data(etf='bbva accion dj eurostoxx 50',
country='spain')
etfs_ger = investpy.get_etfs_list(country='germany')
investpy.get_stock_countries()
#%% get finax tickers
from_date = '01/01/1990'
to_date = '21/03/2021'
search = investpy.search_quotes('FRCK')
df_frck = search[0].retrieve_historical_data(from_date=from_date,
to_date=to_date)
#df_frck.columns = 'frck_' + df_frck.columns
search = investpy.search_quotes('XXSC')
df_xxsc = search[0].retrieve_historical_data(from_date=from_date,
to_date=to_date) # etf - xetra
#df_xxsc.columns = 'xxsc_' + df_xxsc.columns
search = investpy.search_quotes('XSX6')
df_xsx6 = search[1].retrieve_historical_data(from_date=from_date,
to_date=to_date)
#df_xsx6.columns = 'xsx6_' + df_xsx6.columns
search = investpy.search_quotes('XHYA')
#from julia import Man
from scipy.integrate import quad, dblquad
#from pybayes.pdfs import CPdf
from scipy.special import gamma
from arch import arch_model
import pyflux as pf
import sys
import investpy
#Main.include("try3.jl")
#pos=int(input("enter stock pos: "))
#lb = int(input("enter no of returns: "))
#k=pd.read_pickle('/home/sahil/projdir/dailydata.pkl')
#k1=k[k.Symbol==k.Symbol.unique()[pos]].iloc[-lb:]
k1 = investpy.search_quotes(text='AARTIIND',
products=['stocks'],
countries=['India'],
n_results=2)[0].retrieve_historical_data(
from_date='01/01/2014', to_date='07/12/2020')
#k1=investpy.search_quotes(text='AARTIIND',products=['stocks'],countries=['India'],n_results=2)[0].retrieve_historical_data(from_date='01/01/2019',to_date='07/12/2020')
k1['retlog'] = np.log(k1.Close / k1.Open) * 100
k1['ret'] = (k1.Close / k1.Open - 1) * 100
filename = "/home/sahil/pythonbackup/todays_stock1.pkl"
if os.path.exists(filename):
os.remove(filename)
k1.to_pickle(filename)
print("saved")
else:
k1.to_pickle(filename)
#k1.to_pickle("todays_stock1.pkl")
#k1 = pd.read_pickle("todays_stock.pkl")
import investpy
search_results = investpy.search_quotes(text='msci',
products=['stocks'],
countries=['united states'],
n_results=10)
import investpy
res = investpy.search_quotes(text='WSP', countries=['brazil'])
res = res[0] if type(res) == list else res # pick the first one
ret = investpy.get_stock_historical_data(stock=res.symbol,
name=res.name,
country=res.country,
stock_currency='USD',
id_=res.id_,
from_date='01/01/2021',
to_date='13/04/2021',
as_json=True)
print(ret)
def test_investpy_search():
"""
This function checks that investpy search function works properly.
"""
params = [{
'text': 'bbva',
'products': None,
'countries': None,
'n_results': 5
}, {
'text': 'spain 3y',
'products': None,
'countries': None,
'n_results': 5
}, {
'text': 'ibex 35',
'products': None,
'countries': None,
'n_results': 5
}, {
'text': 'bnp daxplus',
'products': None,
'countries': None,
'n_results': None
}, {
'text': 'apple',
'products': ['stocks'],
'countries': ['united states'],
'n_results': 1
}, {
'text': 'apple',
'products': ['stocks'],
'countries': ['united states'],
'n_results': 5
}]
for param in params:
results = investpy.search_quotes(text=param['text'],
products=param['products'],
countries=param['countries'],
n_results=param['n_results'])
dates = [
{
'from_date': '01/01/2018',
'to_date': '01/01/2019'
},
{
'from_date': '01/01/1990',
'to_date': '01/01/2019'
},
]
if isinstance(results, list):
result = results[0]
else:
result = results
print(result)
assert result.retrieve_recent_data() is not None
for date in dates:
assert result.retrieve_historical_data(
from_date=date['from_date'],
to_date=date['to_date']) is not None
assert result.retrieve_currency() is not None
assert result.retrieve_technical_indicators() is not None
financial_products = [('stocks', 'apple'), ('etfs', 'apple'),
('commodities', 'apple'), ('currencies', 'usd'),
('funds', 'apple'), ('bonds', 'apple'),
('cryptos', 'bitcoin'), ('certificates', 'apple'),
('indices', 'apple'), ('fxfutures', 'usd')]
for product_type, product_name in financial_products:
search_result = investpy.search_quotes(text=product_name,
products=[product_type],
n_results=1)
assert search_result.retrieve_information() is not None
import pdb
from mle import *
#import julia
import scipy.optimize as opt
from julia.api import Julia
#j=julia.Julia(compiled_modules=False)
#from julia import Man
from scipy.integrate import quad,dblquad
#from pybayes.pdfs import CPdf
from scipy.special import gamma
from arch import arch_model
import pyflux as pf
import sys
import investpy
import scipy
k1=investpy.search_quotes(text='GAYAPROJ',products=['stocks'],countries=['India'],n_results=2)[0].retrieve_historical_data(from_date='01/01/2018',to_date='16/12/2018')
if path.exists("/home/sahil/pythonbackup/todays_stock.pkl"):
os.remove("/home/sahil/pythonbackup/todays_stock.pkl")
else:
k1.to_pickle("todays_stock.pkl")
k1['dayret'] = ((k1.Close/k1.Open)-1)*100
k1['logdayret'] = np.log((k1.Close/k1.Open))*100
#k1['dayret'] = k1.Close.pct_change()*100
k1 = k1.dropna()
print("long term hold ret")
print(((k1.Close[-1]/k1.Open[0])-1)*100)
print("intraday hold ret")
print((((k1.Close/k1.Open)-1)*100).sum())
print("normal ret skew")
print(scipy.stats.skew(k1.dayret))
import investpy
res = investpy.search_quotes(text='eimi', countries=['united kingdom'])
res = res[0] # pick the first one
ret = investpy.get_stock_historical_data(stock=res.symbol, name=res.name, country=res.country, stock_currency='USD',
id_=res.id_, from_date='01/01/2021', to_date='13/04/2021', as_json=True)
for row_index, row in ret.iteritems():
print(row)
