def test_get_data_interval(self):
# daily interval data
pan = web.get_data_morningstar(symbols='XOM',
start='2013-01-01',
end='2013-12-31',
interval='d')
assert len(pan) == 261
# weekly interval data
pan = web.get_data_morningstar(symbols='XOM',
start='2013-01-01',
end='2013-12-31',
interval='w')
assert len(pan) == 54
# monthly interval data
pan = web.get_data_morningstar(symbols='XOM',
start='2013-01-01',
end='2013-12-31',
interval='m')
assert len(pan) == 13
# test fail on invalid interval
with pytest.raises(ValueError):
web.get_data_morningstar('XOM', interval='NOT VALID')
def incl_dividend_column_multi(self):
df = web.get_data_morningstar(symbols=['XOM', 'MSFT'],
start='2013-01-01',
end='2013-03-04',
incl_dividends=True)
assert ("isDividend" in df)
def test_get_data_multiple_symbols_two_dates(self):
df = web.get_data_morningstar(symbols=['XOM', 'MSFT'],
start='2013-01-01',
end='2013-03-04')
assert len(df.index.levels[0]) == 2
assert 'XOM' in df.index.levels[0]
assert 'MSFT' in df.index.levels[0]
# sanity checking
assert df.dtypes['Close'] == np.float64
assert df.dtypes['Open'] == np.float64
assert df.dtypes['Low'] == np.float64
assert df.dtypes['High'] == np.float64
assert df.dtypes['Volume'] == np.int64
import pickle
import time
style.use('ggplot') #Makes graph look nice
start_day = datetime.date(2013, 1, 1)
end_day = datetime.date(2018, 4, 10)
delta_days = end_day - start_day
days_inbetween = [start_day]
for i in range(delta_days.days + 1):
days_inbetween.append(start_day + datetime.timedelta(days=i))
df = web.get_data_morningstar('AAPL', start_day, end_day)
# df = quandl.get("EOD/HD", authtoken="jMntsFGxviSof9LiXbG2")
df = df[['Open', 'High', 'Low', 'Close',
'Volume']] #Grab important datasets only
old_days_inbetween = list(df.index) #Get all dates in same format
dictionary_days = dict(zip(old_days_inbetween, days_inbetween))
df.rename(index=dictionary_days, inplace=True)
df.reset_index(inplace=True)
df.drop("Symbol", axis=1, inplace=True)
df.set_index('Date', inplace=True) #Replaces index of Symbol with Date
df['HL_PCT'] = 100 * (
(df['High'] - df['Low']) / df['High']) #How volitile the market is
df['PCT_Change'] = 100 * ((df['Close'] - df['Open']) / df['Open']
def test_get_data_multiple_symbols(self):
# just test that we succeed
sl = ['AAPL', 'AMZN', 'GOOG']
web.get_data_morningstar(sl, '2012')
def test_get_data_single_symbol(self):
# single symbol
# just test that we succeed
web.get_data_morningstar('GOOG')
def excl_volume_column_multi(self):
df = web.get_data_morningstar(symbols=["XOM", "MSFT"],
start='2013-01-01',
end='2013-03-04',
incl_volume=False)
assert ("Volume" not in df.keys())
#print(filehandle)
pickle.dump(df, filehandle)
else:
print('Already have KR for {}'.format(ticker))
pass
def create_price_history_table(tickers):
PH = pd.DataFrame()
for i in range(0,len(tickers)):
print tickers[i]
return PH
start = dt.datetime(2017,2,1)
end = dt.datetime.now()
df = web.get_data_morningstar('XJSE:FSR', start, end, incl_dividends = True, incl_volume = True, incl_splits = True)
df['100ma'] = df['Close'].rolling(window=100, min_periods=0).mean()
print(df.head(1000))
def main():
"""main entry poin t for script"""
create_directories()
ignore = np.array(['ANDV', 'BKNG', 'BHF', 'CBRE', 'DWDP', 'DXC','TPR','UAA','WELL'])
#get_price_history(ignoreset = ignore)
#get_key_ratios(ignoreset=ignore)
pass
if __name__ == '__main__':
sys.exit(main())
