def performance_analysis_MF(pnls_or_values,
initial_cash=1,
benchmark='^GSPC',
risk_free='^IRX',
mar=0.0,
input_type='value'):
if input_type == 'value':
values_df = pd.Series(pnls_or_values)
elif input_type == 'pnl':
values_df = pd.Series(pnls_or_values).cumsum() + initial_cash
values_df.index = pd.to_datetime(values_df.index)
start_date = values_df.index[0]
end_date = values_df.index[-1]
# add the initial portfolio values
values_df = pd.concat(
[pd.Series([initial_cash], index=[start_date + BDay(-1)]), values_df])
# calc the daily returns
returns_df = (values_df - values_df.shift(1)) / values_df.shift(1)
returns_df = returns_df.dropna()
# calc the annualized return
cum_return = values_df.iloc[1:] / initial_cash - 1
annual_returns_df = (cum_return +
1)**(252 / np.array(range(1,
len(returns_df) + 1))) - 1
# calc the annualized volatility
annual_vol = returns_df.std() * np.sqrt(252)
# calc the Sharpe ratio / sortino ratio
if risk_free:
# get the risk-free prices
RF_quotes = web.DataReader(risk_free, 'yahoo', start_date + BDay(-1),
end_date)['Close']
# get the expected risk-free rate
risk_free = np.mean(1 / (1 - RF_quotes * 0.01) - 1)
else:
risk_free = 0.0
daily_risk_free = risk_free / 252
daily_mar = mar / 252
sharpe_ratio = (returns_df - daily_risk_free).mean() / (
returns_df - daily_risk_free).std() * 252**0.5
sortino_ratio = (returns_df.mean() - daily_mar) / (
returns_df[returns_df < daily_mar]).std() * 252**0.5
#sharpe_ratio = (returns_df.mean()*252 - risk_free) / ((returns_df - daily_risk_free).std()*252**0.5)
#sortino_ratio = (returns_df.mean()*252 - mar) / ((returns_df[returns_df < daily_mar]).std()*252**0.5)
# calc the maximum drawdown
cum_max_value = (1 + cum_return).cummax()
drawdowns = ((1 + cum_return) - cum_max_value) / cum_max_value
max_drawdown = np.min(drawdowns)
avg_drawdown = drawdowns.mean()
if benchmark:
# get the benchmark prices
benchmark_prices = web.DataReader(benchmark, 'yahoo',
start_date + BDay(-1),
end_date)['Close']
print(benchmark_prices.shape)
# calc the benchmark daily returns
benchmark_returns = (benchmark_prices - benchmark_prices.shift(1)
) / benchmark_prices.shift(1)
benchmark_returns = benchmark_returns.dropna()
# calc the benchmark annualized return
benchmark_cum_return = np.exp(np.log1p(benchmark_returns).cumsum()) - 1
benchmark_annual_returns = (benchmark_cum_return + 1)**(
252 / np.array(range(1,
len(benchmark_returns) + 1))) - 1
# calc the benchmark values based on the same initial_cash of portfolio
benchmark_values = pd.concat([
pd.Series([initial_cash], index=[start_date + BDay(-1)]),
initial_cash * (1 + benchmark_cum_return)
])
# calc the benchmark annualized volatility
benchmark_annual_vol = benchmark_returns.std() * np.sqrt(252)
# calc the maximum drawdown
benchmark_cum_max_value = (1 + benchmark_cum_return).cummax()
benchmark_drawdowns = (
(1 + benchmark_cum_return) -
benchmark_cum_max_value) / benchmark_cum_max_value
benchmark_max_drawdown = np.min(benchmark_drawdowns)
benchmark_avg_drawdown = benchmark_drawdowns.mean()
# compare with the benchmark
relative_return = annual_returns_df.iloc[
-1] - benchmark_annual_returns.iloc[-1]
relative_vol = annual_vol - benchmark_annual_vol
relative_max_drawdown = max_drawdown - benchmark_max_drawdown
relative_avg_drawdown = avg_drawdown - benchmark_avg_drawdown
excess_return_std = (returns_df -
benchmark_returns).std() * np.sqrt(252)
info_ratio = relative_return / excess_return_std
# organize the output
performance = pd.Series()
performance.loc['Begin'] = start_date
performance.loc['End'] = end_date
performance.loc['Duration'] = performance.End - performance.Begin
performance.loc['Initial_Value'] = initial_cash
performance.loc['Highest_Value'] = np.max(values_df)
performance.loc['Lowest_Value'] = np.min(values_df)
performance.loc['Final_Value'] = values_df.iloc[-1]
performance.loc['Total_Return'] = performance['Final_Value'] / performance[
'Initial_Value'] - 1
performance.loc['Total_Return_(Annualized)'] = annual_returns_df.iloc[-1]
performance.loc['Volatility_(Annualized)'] = annual_vol
performance.loc['Max_Drawdown'] = max_drawdown
performance.loc['Avg_Drawdown'] = avg_drawdown
performance.loc['Sharpe_Ratio'] = sharpe_ratio
performance.loc['Sortino_Ratio'] = sortino_ratio
if benchmark:
performance.loc['Relative_Return'] = relative_return
performance.loc['Relative_Vol'] = relative_vol
performance.loc['Relative_Max_DD'] = relative_max_drawdown
performance.loc['Relative_Avg_DD'] = relative_avg_drawdown
performance.loc['Information_Ratio'] = info_ratio
print(performance)
performance.loc['values_data'] = values_df
performance.loc['returns_data'] = returns_df
performance.loc['annual_returns_data'] = annual_returns_df
performance.loc['drawdowns_data'] = drawdowns
if benchmark:
performance.loc['benchmark_values_data'] = benchmark_values
strategy_plot(performance)
return performance
class TestDatetimeIndexOps:
def test_ops_properties_basic(self, datetime_series):
# sanity check that the behavior didn't change
# GH#7206
for op in ["year", "day", "second", "weekday"]:
msg = f"'Series' object has no attribute '{op}'"
with pytest.raises(AttributeError, match=msg):
getattr(datetime_series, op)
# attribute access should still work!
s = Series({"year": 2000, "month": 1, "day": 10})
assert s.year == 2000
assert s.month == 1
assert s.day == 10
msg = "'Series' object has no attribute 'weekday'"
with pytest.raises(AttributeError, match=msg):
s.weekday
def test_repeat_range(self, tz_naive_fixture):
tz = tz_naive_fixture
rng = date_range("1/1/2000", "1/1/2001")
result = rng.repeat(5)
assert result.freq is None
assert len(result) == 5 * len(rng)
index = date_range("2001-01-01", periods=2, freq="D", tz=tz)
exp = DatetimeIndex(
["2001-01-01", "2001-01-01", "2001-01-02", "2001-01-02"], tz=tz)
for res in [index.repeat(2), np.repeat(index, 2)]:
tm.assert_index_equal(res, exp)
assert res.freq is None
index = date_range("2001-01-01", periods=2, freq="2D", tz=tz)
exp = DatetimeIndex(
["2001-01-01", "2001-01-01", "2001-01-03", "2001-01-03"], tz=tz)
for res in [index.repeat(2), np.repeat(index, 2)]:
tm.assert_index_equal(res, exp)
assert res.freq is None
index = DatetimeIndex(["2001-01-01", "NaT", "2003-01-01"], tz=tz)
exp = DatetimeIndex(
[
"2001-01-01",
"2001-01-01",
"2001-01-01",
"NaT",
"NaT",
"NaT",
"2003-01-01",
"2003-01-01",
"2003-01-01",
],
tz=tz,
)
for res in [index.repeat(3), np.repeat(index, 3)]:
tm.assert_index_equal(res, exp)
assert res.freq is None
def test_repeat(self, tz_naive_fixture):
tz = tz_naive_fixture
reps = 2
msg = "the 'axis' parameter is not supported"
rng = date_range(start="2016-01-01", periods=2, freq="30Min", tz=tz)
expected_rng = DatetimeIndex([
Timestamp("2016-01-01 00:00:00", tz=tz, freq="30T"),
Timestamp("2016-01-01 00:00:00", tz=tz, freq="30T"),
Timestamp("2016-01-01 00:30:00", tz=tz, freq="30T"),
Timestamp("2016-01-01 00:30:00", tz=tz, freq="30T"),
])
res = rng.repeat(reps)
tm.assert_index_equal(res, expected_rng)
assert res.freq is None
tm.assert_index_equal(np.repeat(rng, reps), expected_rng)
with pytest.raises(ValueError, match=msg):
np.repeat(rng, reps, axis=1)
@pytest.mark.parametrize(
"freq,expected",
[
("A", "day"),
("Q", "day"),
("M", "day"),
("D", "day"),
("H", "hour"),
("T", "minute"),
("S", "second"),
("L", "millisecond"),
("U", "microsecond"),
],
)
def test_resolution(self, tz_naive_fixture, freq, expected):
tz = tz_naive_fixture
if freq == "A" and not IS64 and isinstance(tz, tzlocal):
pytest.xfail(reason="OverflowError inside tzlocal past 2038")
idx = date_range(start="2013-04-01", periods=30, freq=freq, tz=tz)
assert idx.resolution == expected
def test_value_counts_unique(self, tz_naive_fixture):
tz = tz_naive_fixture
# GH 7735
idx = date_range("2011-01-01 09:00", freq="H", periods=10)
# create repeated values, 'n'th element is repeated by n+1 times
idx = DatetimeIndex(np.repeat(idx.values, range(1,
len(idx) + 1)),
tz=tz)
exp_idx = date_range("2011-01-01 18:00", freq="-1H", periods=10, tz=tz)
expected = Series(range(10, 0, -1), index=exp_idx, dtype="int64")
expected.index = expected.index._with_freq(None)
for obj in [idx, Series(idx)]:
tm.assert_series_equal(obj.value_counts(), expected)
expected = date_range("2011-01-01 09:00", freq="H", periods=10, tz=tz)
expected = expected._with_freq(None)
tm.assert_index_equal(idx.unique(), expected)
idx = DatetimeIndex(
[
"2013-01-01 09:00",
"2013-01-01 09:00",
"2013-01-01 09:00",
"2013-01-01 08:00",
"2013-01-01 08:00",
pd.NaT,
],
tz=tz,
)
exp_idx = DatetimeIndex(["2013-01-01 09:00", "2013-01-01 08:00"],
tz=tz)
expected = Series([3, 2], index=exp_idx)
for obj in [idx, Series(idx)]:
tm.assert_series_equal(obj.value_counts(), expected)
exp_idx = DatetimeIndex(
["2013-01-01 09:00", "2013-01-01 08:00", pd.NaT], tz=tz)
expected = Series([3, 2, 1], index=exp_idx)
for obj in [idx, Series(idx)]:
tm.assert_series_equal(obj.value_counts(dropna=False), expected)
tm.assert_index_equal(idx.unique(), exp_idx)
@pytest.mark.parametrize(
"idx",
[
DatetimeIndex(["2011-01-01", "2011-01-02", "2011-01-03"],
freq="D",
name="idx"),
DatetimeIndex(
["2011-01-01 09:00", "2011-01-01 10:00", "2011-01-01 11:00"],
freq="H",
name="tzidx",
tz="Asia/Tokyo",
),
],
)
def test_order_with_freq(self, idx):
ordered = idx.sort_values()
tm.assert_index_equal(ordered, idx)
assert ordered.freq == idx.freq
ordered = idx.sort_values(ascending=False)
expected = idx[::-1]
tm.assert_index_equal(ordered, expected)
assert ordered.freq == expected.freq
assert ordered.freq.n == -1
ordered, indexer = idx.sort_values(return_indexer=True)
tm.assert_index_equal(ordered, idx)
tm.assert_numpy_array_equal(indexer,
np.array([0, 1, 2]),
check_dtype=False)
assert ordered.freq == idx.freq
ordered, indexer = idx.sort_values(return_indexer=True,
ascending=False)
expected = idx[::-1]
tm.assert_index_equal(ordered, expected)
tm.assert_numpy_array_equal(indexer,
np.array([2, 1, 0]),
check_dtype=False)
assert ordered.freq == expected.freq
assert ordered.freq.n == -1
@pytest.mark.parametrize(
"index_dates,expected_dates",
[
(
[
"2011-01-01", "2011-01-03", "2011-01-05", "2011-01-02",
"2011-01-01"
],
[
"2011-01-01", "2011-01-01", "2011-01-02", "2011-01-03",
"2011-01-05"
],
),
(
[
"2011-01-01", "2011-01-03", "2011-01-05", "2011-01-02",
"2011-01-01"
],
[
"2011-01-01", "2011-01-01", "2011-01-02", "2011-01-03",
"2011-01-05"
],
),
(
[pd.NaT, "2011-01-03", "2011-01-05", "2011-01-02", pd.NaT],
[pd.NaT, pd.NaT, "2011-01-02", "2011-01-03", "2011-01-05"],
),
],
)
def test_order_without_freq(self, index_dates, expected_dates,
tz_naive_fixture):
tz = tz_naive_fixture
# without freq
index = DatetimeIndex(index_dates, tz=tz, name="idx")
expected = DatetimeIndex(expected_dates, tz=tz, name="idx")
ordered = index.sort_values(na_position="first")
tm.assert_index_equal(ordered, expected)
assert ordered.freq is None
ordered = index.sort_values(ascending=False)
tm.assert_index_equal(ordered, expected[::-1])
assert ordered.freq is None
ordered, indexer = index.sort_values(return_indexer=True,
na_position="first")
tm.assert_index_equal(ordered, expected)
exp = np.array([0, 4, 3, 1, 2])
tm.assert_numpy_array_equal(indexer, exp, check_dtype=False)
assert ordered.freq is None
ordered, indexer = index.sort_values(return_indexer=True,
ascending=False)
tm.assert_index_equal(ordered, expected[::-1])
exp = np.array([2, 1, 3, 0, 4])
tm.assert_numpy_array_equal(indexer, exp, check_dtype=False)
assert ordered.freq is None
def test_drop_duplicates_metadata(self, freq_sample):
# GH 10115
idx = date_range("2011-01-01",
freq=freq_sample,
periods=10,
name="idx")
result = idx.drop_duplicates()
tm.assert_index_equal(idx, result)
assert idx.freq == result.freq
idx_dup = idx.append(idx)
assert idx_dup.freq is None # freq is reset
result = idx_dup.drop_duplicates()
expected = idx._with_freq(None)
tm.assert_index_equal(result, expected)
assert result.freq is None
@pytest.mark.parametrize(
"keep, expected, index",
[
("first", np.concatenate(
([False] * 10, [True] * 5)), np.arange(0, 10)),
("last", np.concatenate(
([True] * 5, [False] * 10)), np.arange(5, 15)),
(
False,
np.concatenate(([True] * 5, [False] * 5, [True] * 5)),
np.arange(5, 10),
),
],
)
def test_drop_duplicates(self, freq_sample, keep, expected, index):
# to check Index/Series compat
idx = date_range("2011-01-01",
freq=freq_sample,
periods=10,
name="idx")
idx = idx.append(idx[:5])
tm.assert_numpy_array_equal(idx.duplicated(keep=keep), expected)
expected = idx[~expected]
result = idx.drop_duplicates(keep=keep)
tm.assert_index_equal(result, expected)
result = Series(idx).drop_duplicates(keep=keep)
tm.assert_series_equal(result, Series(expected, index=index))
def test_infer_freq(self, freq_sample):
# GH 11018
idx = date_range("2011-01-01 09:00:00", freq=freq_sample, periods=10)
result = DatetimeIndex(idx.asi8, freq="infer")
tm.assert_index_equal(idx, result)
assert result.freq == freq_sample
def test_nat(self, tz_naive_fixture):
tz = tz_naive_fixture
assert DatetimeIndex._na_value is pd.NaT
assert DatetimeIndex([])._na_value is pd.NaT
idx = DatetimeIndex(["2011-01-01", "2011-01-02"], tz=tz)
assert idx._can_hold_na
tm.assert_numpy_array_equal(idx._isnan, np.array([False, False]))
assert idx.hasnans is False
tm.assert_numpy_array_equal(idx._nan_idxs, np.array([], dtype=np.intp))
idx = DatetimeIndex(["2011-01-01", "NaT"], tz=tz)
assert idx._can_hold_na
tm.assert_numpy_array_equal(idx._isnan, np.array([False, True]))
assert idx.hasnans is True
tm.assert_numpy_array_equal(idx._nan_idxs, np.array([1],
dtype=np.intp))
@pytest.mark.parametrize("values",
[["20180101", "20180103", "20180105"], []])
@pytest.mark.parametrize(
"freq", ["2D", Day(2), "2B",
BDay(2), "48H", Hour(48)])
@pytest.mark.parametrize("tz", [None, "US/Eastern"])
def test_freq_setter(self, values, freq, tz):
# GH 20678
idx = DatetimeIndex(values, tz=tz)
# can set to an offset, converting from string if necessary
idx._data.freq = freq
assert idx.freq == freq
assert isinstance(idx.freq, DateOffset)
# can reset to None
idx._data.freq = None
assert idx.freq is None
def test_freq_setter_errors(self):
# GH 20678
idx = DatetimeIndex(["20180101", "20180103", "20180105"])
# setting with an incompatible freq
msg = ("Inferred frequency 2D from passed values does not conform to "
"passed frequency 5D")
with pytest.raises(ValueError, match=msg):
idx._data.freq = "5D"
# setting with non-freq string
with pytest.raises(ValueError, match="Invalid frequency"):
idx._data.freq = "foo"
def test_freq_view_safe(self):
# Setting the freq for one DatetimeIndex shouldn't alter the freq
# for another that views the same data
dti = date_range("2016-01-01", periods=5)
dta = dti._data
dti2 = DatetimeIndex(dta)._with_freq(None)
assert dti2.freq is None
# Original was not altered
assert dti.freq == "D"
assert dta.freq == "D"
def test_mismatching_tz_raises_err(self, start, end):
# issue 18488
with pytest.raises(TypeError):
pd.date_range(start, end)
with pytest.raises(TypeError):
pd.date_range(start, end, freq=BDay())
# Additionally, can make custom frequncy - see hourly timestamp range below
pd.date_range('2015-07-03', periods=8, freq='H')
# or, a sequence of durations increasing by an hour:
pd.timedelta_range(0, periods=10, freq='H')
### Frequencies and Offsets
# (table of codes)
# stuff ------
pd.timedelta_range(0, periods=9, freq="2H30T")
# get range of 5 business days, beginning on 2015-07-01
from pandas.tseries.offsets import BDay
pd.date_range('2015-07-01', periods=5, freq=BDay())
### Resampling, Shifting, and Windowing
from pandas_datareader import data
goog = data.DataReader('GOOG', start='2004', end='2016', data_source='google')
# NOTE: "ImmediateDeprecationError - Google Finance dep. due to API breaks"
# will not be able to complete section notes. Goes into some basic plots.
## Resampling and Converting Frequencies
# still uses deprecated functionality
### Example: Visualizing Seattle Bicycle Counts
# data get:
startingYear = 2015 # First year of data
endingYear = 2015 # Last year of data
for year in range(startingYear, endingYear + 1):
makeDirectory(year)
startTimeAcquireData = time()
rawData = {year: {}}
if year >= 1999:
firstDateOfYear = datetime.strptime('01/01/{:d}'.format(year),
'%m/%d/%Y')
daysUntilStartingDay = random.randint(0, 5)
recordDate = firstDateOfYear + BDay(daysUntilStartingDay)
while recordDate.year == year:
recordURL = getCongressionalRecordURL(year, date=recordDate)
try:
filePath = downloadPDFFile(recordURL, year, date=recordDate)
textPageDict = parsePDFFile(filePath, everyNPages=7)
recordDateString = str(recordDate)[:10]
rawData[year][recordDateString] = textPageDict
daysUntilNextRecordDate = random.randint(4, 9)
except (PDFSyntaxError, FileNotFoundError):
print('{:s} doesnt have data'.format(str(recordDate)))
daysUntilNextRecordDate = random.randint(1, 2)
the date(s) to be converted
Returns
-------
same type as input
date(s) converted to UTC
"""
dt = pd.to_datetime(dt)
try:
dt = dt.tz_localize('UTC')
except TypeError:
dt = dt.tz_convert('UTC')
return dt
_1_bday = BDay()
def _1_bday_ago():
return pd.Timestamp.now().normalize() - _1_bday
# @deprecated(msg=DATAREADER_DEPRECATION_WARNING)
# def get_fama_french():
# """
# Retrieve Fama-French factors via pandas-datareader
# Returns
# -------
# pandas.DataFrame
# Percent change of Fama-French factors
# """
def test_shift(self):
shifted = self.ts.shift(1)
unshifted = shifted.shift(-1)
tm.assert_index_equal(shifted.index, self.ts.index)
tm.assert_index_equal(unshifted.index, self.ts.index)
tm.assert_numpy_array_equal(unshifted.valid().values,
self.ts.values[:-1])
offset = BDay()
shifted = self.ts.shift(1, freq=offset)
unshifted = shifted.shift(-1, freq=offset)
assert_series_equal(unshifted, self.ts)
unshifted = self.ts.shift(0, freq=offset)
assert_series_equal(unshifted, self.ts)
shifted = self.ts.shift(1, freq='B')
unshifted = shifted.shift(-1, freq='B')
assert_series_equal(unshifted, self.ts)
# corner case
unshifted = self.ts.shift(0)
assert_series_equal(unshifted, self.ts)
# Shifting with PeriodIndex
ps = tm.makePeriodSeries()
shifted = ps.shift(1)
unshifted = shifted.shift(-1)
tm.assert_index_equal(shifted.index, ps.index)
tm.assert_index_equal(unshifted.index, ps.index)
tm.assert_numpy_array_equal(unshifted.valid().values, ps.values[:-1])
shifted2 = ps.shift(1, 'B')
shifted3 = ps.shift(1, BDay())
assert_series_equal(shifted2, shifted3)
assert_series_equal(ps, shifted2.shift(-1, 'B'))
pytest.raises(ValueError, ps.shift, freq='D')
# legacy support
shifted4 = ps.shift(1, freq='B')
assert_series_equal(shifted2, shifted4)
shifted5 = ps.shift(1, freq=BDay())
assert_series_equal(shifted5, shifted4)
# 32-bit taking
# GH 8129
index = date_range('2000-01-01', periods=5)
for dtype in ['int32', 'int64']:
s1 = Series(np.arange(5, dtype=dtype), index=index)
p = s1.iloc[1]
result = s1.shift(periods=p)
expected = Series([np.nan, 0, 1, 2, 3], index=index)
assert_series_equal(result, expected)
# xref 8260
# with tz
s = Series(date_range('2000-01-01 09:00:00',
periods=5,
tz='US/Eastern'),
name='foo')
result = s - s.shift()
exp = Series(TimedeltaIndex(['NaT'] + ['1 days'] * 4), name='foo')
assert_series_equal(result, exp)
# incompat tz
s2 = Series(date_range('2000-01-01 09:00:00', periods=5, tz='CET'),
name='foo')
pytest.raises(ValueError, lambda: s - s2)
NORM_FACTOR = 100
# Matching delay
MAX_LAG = pd.Timedelta('1 hours')
MATCH_TOL_DICT = {
'1H': pd.Timedelta('1 hours'),
'0.5H': pd.Timedelta('0.5 hours'),
'0.1H': pd.Timedelta('0.1 hours')
}
MATCH_TOL = '0.1H'
# Hedging frequency
OFFSET_DICT = {
'1H': [pd.Timedelta('1 hours'), '_1H'],
'1D': [BDay(1), '_1D'],
'2D': [BDay(2), '_2D']
}
FREQ = '1H'
T_FIRSTHALF = datetime.time(8, 30)
if FREQ == '1H':
DT = 0.
T_LASTHALF = datetime.time(15, 15)
if FREQ == '1D':
DT = 1. / 253.
T_LASTHALF = datetime.time(16, 0)
if FREQ == '2D':
DT = 2. / 253.
T_LASTHALF = datetime.time(16, 0)
def main():
temptime = datetime.datetime.now()
print(temptime)
warnings.filterwarnings("error")
#final df
wtf = pd.DataFrame()
final = pd.DataFrame()
#unprocessed df
unproc = pd.DataFrame()
unproc['complist'] = ''
#number of unprocessed comp
pos = 0
#define date where we start fetching data, or check if any data exists
try:
maindf = pd.read_csv(os.path.join('allprice.csv'))
final['Date'] = maindf['Date']
day_num = len(maindf['Date'])
start = datetime.datetime.strptime(str(maindf.loc[day_num - 1][0]),
'%Y-%m-%d')
end = datetime.datetime.today()
status = "update"
day = day_num
for x in perdelta(start, end, timedelta(days=1)):
final.set_value(day, 'Date', x)
day += 1
except (UnboundLocalError, OSError):
start = datetime.datetime(2010, 1, 1)
end = datetime.datetime.today()
day_num = 0
maindf = pd.DataFrame()
status = "new"
complist = pd.read_csv(path.join('Data/companylist.csv'))
#traverse through the company list
for x in range(len(complist)):
time = datetime.datetime.now()
day_num1 = day_num
rsdf = pd.DataFrame()
name = complist.ix[x]
file_num = 1
#fetch data
try:
data = web.DataReader(complist.ix[x], 'yahoo', start, end)
print(complist.ix[x] + "'s Data obtained from Yahoo " +
str(time.isoformat()))
except:
try:
data = web.DataReader(complist.ix[x], 'google', start, end)
print(complist.ix[x] + "'s Data obtained from Google " +
str(time.isoformat()))
except:
try:
data = web.DataReader(complist.ix[x], 'fred', start, end)
print(complist.ix[x] + "'s Data obtained from Fred " +
str(time.isoformat()))
except:
continue
continue
print(complist.ix[x] + "Cannot be Obtained " +
str(time.isoformat()))
continue
#create or append base on the status
if status == "new":
rsdf[name] = data['Adj Close']
else:
try:
rsdf[name] = maindf[name]
rsdf.columns.values[0] = 'Adj Close'
except KeyError:
unproc.set_value(pos, 'complist', str(name))
pos += 1
comprice = data['Adj Close']
if (status == "update"):
rsdf = rsdf.append(comprice, ignore_index=True)
rsdf.rename(columns={'Adj Close': str(name)}, inplace=True)
if (status == 'new' and x == 0):
#since it's new, add an index Date column from 2010 to today
date = pd.date_range('20100101', end, freq=BDay())
final = pd.DataFrame(rsdf[name], index=date)
else:
#dump the data for every 1000 companies to decrease processing time
if (x % 1000 == 0 and x != 0):
if file_num != 1:
final.to_csv(path.join('all_price' + str(file_num) +
'.csv'),
index=False)
else:
final.to_csv(path + 'all_price' + str(file_num) + '.csv',
index=True)
file_num += 1
final = pd.DataFrame(rsdf[name])
else:
final = final.join(rsdf[name])
#dump all data frame to file
final.to_csv(path + 'all_price' + str(file_num) + '.csv', index=False)
y = 1
final = merge_file(y, file_num, path)
final.rename(columns={'Unnamed: 0': 'Date'}, inplace=True)
temptime = datetime.datetime.now()
final.to_csv(path.join('all_price.csv'), index=False)
#list of unprocessed companies due to not found data
unproc.to_csv(path.join('unproc.csv'), index=False)
print(temptime)
def test_shift(self, datetime_series):
shifted = datetime_series.shift(1)
unshifted = shifted.shift(-1)
tm.assert_index_equal(shifted.index, datetime_series.index)
tm.assert_index_equal(unshifted.index, datetime_series.index)
tm.assert_numpy_array_equal(
unshifted.dropna().values, datetime_series.values[:-1]
)
offset = BDay()
shifted = datetime_series.shift(1, freq=offset)
unshifted = shifted.shift(-1, freq=offset)
tm.assert_series_equal(unshifted, datetime_series)
unshifted = datetime_series.shift(0, freq=offset)
tm.assert_series_equal(unshifted, datetime_series)
shifted = datetime_series.shift(1, freq="B")
unshifted = shifted.shift(-1, freq="B")
tm.assert_series_equal(unshifted, datetime_series)
# corner case
unshifted = datetime_series.shift(0)
tm.assert_series_equal(unshifted, datetime_series)
# Shifting with PeriodIndex
ps = tm.makePeriodSeries()
shifted = ps.shift(1)
unshifted = shifted.shift(-1)
tm.assert_index_equal(shifted.index, ps.index)
tm.assert_index_equal(unshifted.index, ps.index)
tm.assert_numpy_array_equal(unshifted.dropna().values, ps.values[:-1])
shifted2 = ps.shift(1, "B")
shifted3 = ps.shift(1, BDay())
tm.assert_series_equal(shifted2, shifted3)
tm.assert_series_equal(ps, shifted2.shift(-1, "B"))
msg = "Given freq D does not match PeriodIndex freq B"
with pytest.raises(ValueError, match=msg):
ps.shift(freq="D")
# legacy support
shifted4 = ps.shift(1, freq="B")
tm.assert_series_equal(shifted2, shifted4)
shifted5 = ps.shift(1, freq=BDay())
tm.assert_series_equal(shifted5, shifted4)
# 32-bit taking
# GH 8129
index = date_range("2000-01-01", periods=5)
for dtype in ["int32", "int64"]:
s1 = Series(np.arange(5, dtype=dtype), index=index)
p = s1.iloc[1]
result = s1.shift(periods=p)
expected = Series([np.nan, 0, 1, 2, 3], index=index)
tm.assert_series_equal(result, expected)
# xref 8260
# with tz
s = Series(
date_range("2000-01-01 09:00:00", periods=5, tz="US/Eastern"), name="foo"
)
result = s - s.shift()
exp = Series(TimedeltaIndex(["NaT"] + ["1 days"] * 4), name="foo")
tm.assert_series_equal(result, exp)
# incompat tz
s2 = Series(date_range("2000-01-01 09:00:00", periods=5, tz="CET"), name="foo")
msg = "DatetimeArray subtraction must have the same timezones or no timezones"
with pytest.raises(TypeError, match=msg):
s - s2
def freight_and_quality_exceptions():
if crude in ('Forties'):
df_freight['Buzzard_Content'] = forties_sulphur[
'BuzzardContent']
df_freight['Implied_Sulphur'] = df_freight[
'Buzzard_Content'] * 0.012 + 0.003
df_freight['De-Escalator_Threshold'] = np.round(
df_freight['Implied_Sulphur'], 3)
df_freight['De-Escalator_Counts'] = np.minimum(
0, 6 - df_freight['Implied_Sulphur'] * 1000)
df_freight['Platts_De_Esc'] = total['AAUXL00']
df_freight['Forties_Margin_Impact'] = df_freight[
'Platts_De_Esc'] * df_freight['De-Escalator_Counts'] * -1
df_freight['Costs'] += df_freight['Forties_Margin_Impact']
if crude in ('Basrah Light', 'Basrah Heavy'):
"""This handles the freight escalation calculation from Iraq - the base is sent by SOMO, and table is in databse / excel wb"""
monthly_averages = total['PFAOH00'].asfreq(BDay(
)).resample('BMS').mean(
) # resampled so we have the business month start, corrects averaging error if cma
func_ma_on_days = lambda x: (monthly_averages.loc[
(monthly_averages.index.month == x.month) &
(monthly_averages.index.year == x.year)]).iat[0]
"""Create funcs to handle basrah base and flat rate values, apply over df and calc esclator"""
func_ws_base = lambda x: (basrah_ws_base.loc[
(basrah_ws_base.index.year == x.year)]['SOMO_WS']).iat[0]
func_fr = lambda x: (basrah_ws_base.loc[
(basrah_ws_base.index.year == x.year)]['SOMO_FlatRate']
).iat[0]
func_bhapi = lambda x: (basrah_ws_base.loc[
(basrah_ws_base.index.year == x.year)]['BasrahHeavyAPI']
).iat[0]
func_blapi = lambda x: (basrah_ws_base.loc[
(basrah_ws_base.index.year == x.year)]['BasrahLightAPI']
).iat[0]
df_freight['Date'] = df_freight.index
df_freight['WS Month Avg'] = df_freight['Date'].apply(
func_ma_on_days)
df_freight['SOMO Base WS'] = df_freight['Date'].apply(
func_ws_base)
# We have to apply the corrcetion here after SOMO dropped their base rate earlier this year - assumption
# only valid for 2018
df_freight['SOMO Base WS'].iloc[
(df_freight.index >= dt(2018, 4, 1))
& (df_freight.index <= dt(2018, 12, 31))] = 25
df_freight['Base_FR_for_esc'] = df_freight['Date'].apply(
func_fr)
if crude == 'Basrah Light':
df_freight['API Esc'] = df_freight['Date'].apply(
func_blapi)
else:
df_freight['API Esc'] = df_freight['Date'].apply(
func_bhapi)
df_freight['WS for Esc'] = (
df_freight['WS Month Avg'] - df_freight['SOMO Base WS']
) * df_freight['Base_FR_for_esc'] / 7.3 / 100
df_freight.drop(['Date'], axis=1, inplace=True)
# South Korean particulars
if ports[ports['PortName'] ==
destination]['Country'].iat[0] == 'South Korea':
# Freight rebate on imported crudes
df_freight['Murban_Freight_Comp'] = total[
'PFAOC00'] / 100 * df_freight[
'Murban_Sing_Flat'] / 7.66 #Murban density conversion
df_freight[
'UKC-Yosu_VLCC'] = total['AASLA00'] * 1000000 / 2000000
df_freight['Freight_Rebate'] = np.maximum(
df_freight['UKC-Yosu_VLCC'] -
df_freight['Murban_Freight_Comp'], 0.6)
df_freight['Costs'] -= df_freight['Freight_Rebate']
# Tax rebate on crudes out of Europe
if ports[ports['PortName'] ==
loadport]['RegionName'].iat[0] in ([
'NW EUROPE', 'MED'
]):
df_freight['FTA_Tax_Rebate'] = 0.006 * total['LCOc1']
df_freight['Costs'] -= df_freight['FTA_Tax_Rebate']
# Tax rebate on crudes out of the US
if ports[ports['PortName'] ==
loadport]['RegionName'].iat[0] in (['N AMERICA']):
df_freight['FTA_Tax_Rebate'] = 0.005 * total['CLc1']
df_freight['Costs'] -= df_freight['FTA_Tax_Rebate']
# Costs ascociated with lifting CPC based on delays
if crude == 'CPC Blend':
df_freight['TS_Delays'] = np.maximum(
total['AAWIL00'] + total['AAWIK00'] - 2, 0)
df_freight['TS_Demur'] = total['AAPED00']
df_freight['TS_Demur_Costs'] = df_freight['TS_Delays'].mul(
df_freight['TS_Demur']) / 130
df_freight['Costs'] += df_freight['TS_Demur_Costs']
# Costs ascociated with lifting Urals, actually a rebate as giving back port costs that are included in CIF price
if crude in (['Urals Nth', 'Urals Med']):
df_freight['Urals_Cif_Rebate'] = 0.11
df_freight['Costs'] -= df_freight['Urals_Discharge_Costs']
if crude == 'Forties':
df_freight['Forties_Mkt_Discount'] = 0.5
df_freight['Costs'] -= df_freight['Forties_Mkt_Discount']
else:
pass
return df_freight
def positions(weights, period, freq=None):
"""
Builds net position values time series, the portfolio percentage invested
in each position.
Parameters
----------
weights: pd.Series
pd.Series containing factor weights, the index contains timestamps at
which the trades are computed and the values correspond to assets
weights
- see factor_weights for more details
period: pandas.Timedelta or string
Assets holding period (1 day, 2 mins, 3 hours etc). It can be a
Timedelta or a string in the format accepted by Timedelta constructor
('1 days', '1D', '30m', '3h', '1D1h', etc)
freq : pandas DateOffset, optional
Used to specify a particular trading calendar. If not present
weights.index.freq will be used
Returns
-------
pd.DataFrame
Assets positions series, datetime on index, assets on columns.
"""
weights = weights.unstack()
if not isinstance(period, pd.Timedelta):
period = pd.Timedelta(period)
if freq is None:
freq = weights.index.freq
if freq is None:
freq = BDay()
warnings.warn("'freq' not set, using business day calendar",
UserWarning)
#
# weights index contains factor computation timestamps, then add returns
# timestamps too (factor timestamps + period) and save them to 'full_idx'
# 'full_idx' index will contain an entry for each point in time the weights
# change and hence they have to be re-computed
#
trades_idx = weights.index.copy()
returns_idx = utils.add_custom_calendar_timedelta(trades_idx, period, freq)
weights_idx = trades_idx.union(returns_idx)
#
# Compute portfolio weights for each point in time contained in the index
#
portfolio_weights = pd.DataFrame(index=weights_idx,
columns=weights.columns)
active_weights = []
for curr_time in weights_idx:
#
# fetch new weights that become available at curr_time and store them
# in active weights
#
if curr_time in weights.index:
assets_weights = weights.loc[curr_time]
expire_ts = utils.add_custom_calendar_timedelta(
curr_time, period, freq)
active_weights.append((expire_ts, assets_weights))
#
# remove expired entry in active_weights (older than 'period')
#
if active_weights:
expire_ts, assets_weights = active_weights[0]
if expire_ts <= curr_time:
active_weights.pop(0)
if not active_weights:
continue
#
# Compute total weights for curr_time and store them
#
tot_weights = [w for (ts, w) in active_weights]
tot_weights = pd.concat(tot_weights, axis=1)
tot_weights = tot_weights.sum(axis=1)
tot_weights /= tot_weights.abs().sum()
portfolio_weights.loc[curr_time] = tot_weights
return portfolio_weights.fillna(0)
def cumulative_returns(returns, period, freq=None):
"""
Builds cumulative returns from 'period' returns. This function simulate the
cumulative effect that a series of gains or losses (the 'retuns') have on
an original amount of capital over a period of time.
if F is the frequency at which returns are computed (e.g. 1 day if
'returns' contains daily values) and N is the period for which the retuns
are computed (e.g. returns after 1 day, 5 hours or 3 days) then:
- if N <= F the cumulative retuns are trivially computed as Compound Return
- if N > F (e.g. F 1 day, and N is 3 days) then the returns overlap and the
cumulative returns are computed building and averaging N interleaved sub
portfolios (started at subsequent periods 1,2,..,N) each one rebalancing
every N periods. This correspond to an algorithm which trades the factor
every single time it is computed, which is statistically more robust and
with a lower volatity compared to an algorithm that trades the factor
every N periods and whose returns depend on the specific starting day of
trading.
Also note that when the factor is not computed at a specific frequency, for
exaple a factor representing a random event, it is not efficient to create
multiples sub-portfolios as it is not certain when the factor will be
traded and this would result in an underleveraged portfolio. In this case
the simulated portfolio is fully invested whenever an event happens and if
a subsequent event occur while the portfolio is still invested in a
previous event then the portfolio is rebalanced and split equally among the
active events.
Parameters
----------
returns: pd.Series
pd.Series containing factor 'period' forward returns, the index
contains timestamps at which the trades are computed and the values
correspond to returns after 'period' time
period: pandas.Timedelta or string
Length of period for which the returns are computed (1 day, 2 mins,
3 hours etc). It can be a Timedelta or a string in the format accepted
by Timedelta constructor ('1 days', '1D', '30m', '3h', '1D1h', etc)
freq : pandas DateOffset, optional
Used to specify a particular trading calendar. If not present
returns.index.freq will be used
Returns
-------
Cumulative returns series : pd.Series
"""
if not isinstance(period, pd.Timedelta):
period = pd.Timedelta(period)
if freq is None:
freq = returns.index.freq
if freq is None:
freq = BDay()
warnings.warn("'freq' not set, using business day calendar",
UserWarning)
#
# returns index contains factor computation timestamps, then add returns
# timestamps too (factor timestamps + period) and save them to 'full_idx'
# Cumulative returns will use 'full_idx' index,because we want a cumulative
# returns value for each entry in 'full_idx'
#
trades_idx = returns.index.copy()
returns_idx = utils.add_custom_calendar_timedelta(trades_idx, period, freq)
full_idx = trades_idx.union(returns_idx)
#
# Build N sub_returns from the single returns Series. Each sub_retuns
# stream will contain non overlapping retuns.
# In the next step we'll compute the portfolio returns averaging the
# returns happening on those overlapping returns streams
#
sub_returns = []
while len(trades_idx) > 0:
#
# select non-overlapping returns starting with first timestamp in index
#
sub_index = []
next = trades_idx.min()
while next <= trades_idx.max():
sub_index.append(next)
next = utils.add_custom_calendar_timedelta(next, period, freq)
# make sure to fetch the next available entry after 'period'
try:
i = trades_idx.get_loc(next, method='bfill')
next = trades_idx[i]
except KeyError:
break
sub_index = pd.DatetimeIndex(sub_index, tz=full_idx.tz)
subret = returns[sub_index]
# make the index to have all entries in 'full_idx'
subret = subret.reindex(full_idx)
#
# compute intermediate returns values for each index in subret that are
# in between the timestaps at which the factors are computed and the
# timestamps at which the 'period' returns actually happen
#
for pret_idx in reversed(sub_index):
pret = subret[pret_idx]
# get all timestamps between factor computation and period returns
pret_end_idx = \
utils.add_custom_calendar_timedelta(pret_idx, period, freq)
slice = subret[(subret.index > pret_idx)
& (subret.index <= pret_end_idx)].index
if pd.isnull(pret):
continue
def rate_of_returns(ret, period):
return ((np.nansum(ret) + 1)**(1. / period)) - 1
# compute intermediate 'period' returns values, note that this also
# moves the final 'period' returns value from trading timestamp to
# trading timestamp + 'period'
for slice_idx in slice:
sub_period = utils.diff_custom_calendar_timedeltas(
pret_idx, slice_idx, freq)
subret[slice_idx] = rate_of_returns(pret, period / sub_period)
subret[pret_idx] = np.nan
# transform returns as percentage change from previous value
subret[slice[1:]] = (subret[slice] + 1).pct_change()[slice[1:]]
sub_returns.append(subret)
trades_idx = trades_idx.difference(sub_index)
#
# Compute portfolio cumulative returns averaging the returns happening on
# overlapping returns streams. Please note that the below algorithm keeps
# into consideration the scenario where a factor is not computed at a fixed
# frequency (e.g. every day) and consequently the returns appears randomly
#
sub_portfolios = pd.concat(sub_returns, axis=1)
portfolio = pd.Series(index=sub_portfolios.index)
for i, (index, row) in enumerate(sub_portfolios.iterrows()):
# check the active portfolios, count() returns non-nans elements
active_subfolios = row.count()
# fill forward portfolio value
portfolio.iloc[i] = portfolio.iloc[i - 1] if i > 0 else 1.
if active_subfolios <= 0:
continue
# current portfolio is the average of active sub_portfolios
portfolio.iloc[i] *= (row + 1).mean(skipna=True)
return portfolio
def performance_analysis_HF(pnls_or_values,
initial_cash=1,
benchmark='^GSPC',
risk_free='^IRX',
mar=0.0,
input_type='value'):
if input_type == 'value':
values_df = pd.Series(pnls_or_values)
elif input_type == 'pnl':
values_df = pd.Series(pnls_or_values).cumsum() + initial_cash
values_df.index = pd.to_datetime(values_df.index)
start_time = values_df.index[0]
end_time = values_df.index[-1]
# downsample the series into 3 min bins
values_df_3T = values_df.resample('3T', label='right').last()
snapshot_per_day = (16 - 9.5) * 60 / 3
# add the initial portfolio values
values_df = pd.concat([
pd.Series([initial_cash], index=[start_time + Minute(-3)]),
values_df_3T
])
# calc the 3-minute returns
returns_df = (values_df - values_df.shift(1)) / values_df.shift(1)
returns_df = returns_df.dropna()
# calc the daily return
cum_return = values_df.iloc[1:] / initial_cash - 1
annual_returns_df = (cum_return + 1)**(
252 * snapshot_per_day / np.array(range(1,
len(returns_df) + 1))) - 1
# calc the daily volatility
annual_vol = returns_df.std() * np.sqrt(252 * snapshot_per_day)
# calc the Sharpe ratio / sortino ratio
if risk_free:
# get the risk-free prices
RF_quotes = web.DataReader(risk_free, 'yahoo', start_time + Minute(-3),
end_time)['Close']
# get the expected risk-free rate
risk_free = np.mean(1 / (1 - RF_quotes * 0.01) - 1)
else:
risk_free = 0.0
# calc the Sharpe ratio / sortino ratio
risk_free_per_snapshot = risk_free / 252 / snapshot_per_day
mar_per_snapshot = mar / 252 / snapshot_per_day
sharpe_ratio = (returns_df.mean() - risk_free_per_snapshot) / (returns_df - risk_free_per_snapshot).std() \
* (252*snapshot_per_day)**0.5
sortino_ratio = (returns_df.mean() - mar_per_snapshot) / (returns_df[returns_df < mar_per_snapshot]).std() \
* (252*snapshot_per_day)**0.5
# calc the maximum drawdown
cum_max_value = (1 + cum_return).cummax()
drawdowns = ((1 + cum_return) - cum_max_value) / cum_max_value
max_drawdown = np.min(drawdowns)
avg_drawdown = drawdowns.mean()
if benchmark:
start_time = values_df.index[0].replace(
tzinfo=timezone('America/New_York'))
end_time = values_df.index[-1].replace(
tzinfo=timezone('America/New_York'))
# get the benchmark prices
benchmark_prices = yf.download(benchmark,
start=start_time.strftime('%Y-%m-%e'),
end=(end_time +
BDay(1)).strftime('%Y-%m-%e'),
interval="1m")['Close']
benchmark_prices_3T = benchmark_prices.resample(
'3T', label='right').last()[values_df_3T.index]
benchmark_prices = pd.concat([
pd.Series([benchmark_prices.iloc[-1]],
index=[start_time + Minute(-3)]), benchmark_prices_3T
])
# calc the benchmark daily returns
benchmark_returns = (benchmark_prices - benchmark_prices.shift(1)
) / benchmark_prices.shift(1)
benchmark_returns = benchmark_returns.dropna()
# calc the benchmark daily return
benchmark_cum_return = np.exp(np.log1p(benchmark_returns).cumsum()) - 1
benchmark_annual_returns = (benchmark_cum_return + 1)**(
252 * snapshot_per_day /
np.array(range(1,
len(benchmark_cum_return) + 1))) - 1
# calc the benchmark values based on the same initial_cash of portfolio
benchmark_values = pd.concat([
pd.Series([initial_cash], index=[start_time + Minute(-3)]),
initial_cash * (1 + benchmark_cum_return)
])
benchmark_values.index = [
local_time.replace(tzinfo=timezone('UTC'))
for local_time in benchmark_values.index
]
# calc the benchmark daily volatility
benchmark_annual_vol = benchmark_returns.std() * np.sqrt(
252 * snapshot_per_day)
# calc the maximum drawdown
benchmark_cum_max_value = (1 + benchmark_cum_return).cummax()
benchmark_drawdowns = (
(1 + benchmark_cum_return) -
benchmark_cum_max_value) / benchmark_cum_max_value
benchmark_max_drawdown = np.min(benchmark_drawdowns)
benchmark_avg_drawdown = benchmark_drawdowns.mean()
# compare with the benchmark
relative_return = annual_returns_df.iloc[
-1] - benchmark_annual_returns.iloc[-1]
relative_vol = annual_vol - benchmark_annual_vol
relative_max_drawdown = max_drawdown - benchmark_max_drawdown
relative_avg_drawdown = avg_drawdown - benchmark_avg_drawdown
excess_return_std = (returns_df - benchmark_returns).std() * np.sqrt(
252 * snapshot_per_day)
info_ratio = relative_return / excess_return_std
# organize the output
performance = pd.Series()
performance.loc['Begin'] = start_time
performance.loc['End'] = end_time
performance.loc['Duration'] = performance.End - performance.Begin
performance.loc['Initial_Value'] = initial_cash
performance.loc['Highest_Value'] = np.max(values_df)
performance.loc['Lowest_Value'] = np.min(values_df)
performance.loc['Final_Value'] = values_df.iloc[-1]
performance.loc['Total_Return'] = performance['Final_Value'] / performance[
'Initial_Value'] - 1
performance.loc['Total_Return_(Annual)'] = annual_returns_df.iloc[-1]
performance.loc['Volatility_(Annual)'] = annual_vol
performance.loc['Max_Drawdown'] = max_drawdown
performance.loc['Avg_Drawdown'] = avg_drawdown
performance.loc['Sharpe_Ratio_(Annual)'] = sharpe_ratio
performance.loc['Sortino_Ratio_(Annual)'] = sortino_ratio
if benchmark:
performance.loc['Relative_Return_(Annual)'] = relative_return
performance.loc['Relative_Vol_(Annual)'] = relative_vol
performance.loc['Relative_Max_DD'] = relative_max_drawdown
performance.loc['Relative_Avg_DD'] = relative_avg_drawdown
performance.loc['Information_Ratio_(Annual)'] = info_ratio
print(performance)
values_df.index = pd.Series([
local_time.replace(tzinfo=timezone('UTC'))
for local_time in values_df.index
])
drawdowns.index = pd.Series([
local_time.replace(tzinfo=timezone('UTC'))
for local_time in drawdowns.index
])
performance.loc['values_data'] = values_df
performance.loc['returns_data'] = returns_df
performance.loc['annual_returns_data'] = annual_returns_df
performance.loc['drawdowns_data'] = drawdowns
if benchmark:
performance.loc['benchmark_values_data'] = benchmark_values
strategy_plot(performance, benchmark, freq='intraday')
return performance
class TestDatetimeIndexOps(Ops):
def setup_method(self, method):
super(TestDatetimeIndexOps, self).setup_method(method)
mask = lambda x: (isinstance(x, DatetimeIndex) or isinstance(
x, PeriodIndex))
self.is_valid_objs = [o for o in self.objs if mask(o)]
self.not_valid_objs = [o for o in self.objs if not mask(o)]
def test_ops_properties(self):
f = lambda x: isinstance(x, DatetimeIndex)
self.check_ops_properties(DatetimeIndex._field_ops, f)
self.check_ops_properties(DatetimeIndex._object_ops, f)
self.check_ops_properties(DatetimeIndex._bool_ops, f)
def test_ops_properties_basic(self):
# sanity check that the behavior didn't change
# GH7206
for op in ['year', 'day', 'second', 'weekday']:
pytest.raises(TypeError, lambda x: getattr(self.dt_series, op))
# attribute access should still work!
s = Series(dict(year=2000, month=1, day=10))
assert s.year == 2000
assert s.month == 1
assert s.day == 10
pytest.raises(AttributeError, lambda: s.weekday)
def test_minmax_tz(self, tz_fixture):
tz = tz_fixture
# monotonic
idx1 = pd.DatetimeIndex(['2011-01-01', '2011-01-02', '2011-01-03'],
tz=tz)
assert idx1.is_monotonic
# non-monotonic
idx2 = pd.DatetimeIndex(
['2011-01-01', pd.NaT, '2011-01-03', '2011-01-02', pd.NaT], tz=tz)
assert not idx2.is_monotonic
for idx in [idx1, idx2]:
assert idx.min() == Timestamp('2011-01-01', tz=tz)
assert idx.max() == Timestamp('2011-01-03', tz=tz)
assert idx.argmin() == 0
assert idx.argmax() == 2
@pytest.mark.parametrize('op', ['min', 'max'])
def test_minmax_nat(self, op):
# Return NaT
obj = DatetimeIndex([])
assert pd.isna(getattr(obj, op)())
obj = DatetimeIndex([pd.NaT])
assert pd.isna(getattr(obj, op)())
obj = DatetimeIndex([pd.NaT, pd.NaT, pd.NaT])
assert pd.isna(getattr(obj, op)())
def test_numpy_minmax(self):
dr = pd.date_range(start='2016-01-15', end='2016-01-20')
assert np.min(dr) == Timestamp('2016-01-15 00:00:00', freq='D')
assert np.max(dr) == Timestamp('2016-01-20 00:00:00', freq='D')
errmsg = "the 'out' parameter is not supported"
tm.assert_raises_regex(ValueError, errmsg, np.min, dr, out=0)
tm.assert_raises_regex(ValueError, errmsg, np.max, dr, out=0)
assert np.argmin(dr) == 0
assert np.argmax(dr) == 5
if not _np_version_under1p10:
errmsg = "the 'out' parameter is not supported"
tm.assert_raises_regex(ValueError, errmsg, np.argmin, dr, out=0)
tm.assert_raises_regex(ValueError, errmsg, np.argmax, dr, out=0)
def test_repeat_range(self, tz_fixture):
tz = tz_fixture
rng = date_range('1/1/2000', '1/1/2001')
result = rng.repeat(5)
assert result.freq is None
assert len(result) == 5 * len(rng)
index = pd.date_range('2001-01-01', periods=2, freq='D', tz=tz)
exp = pd.DatetimeIndex(
['2001-01-01', '2001-01-01', '2001-01-02', '2001-01-02'], tz=tz)
for res in [index.repeat(2), np.repeat(index, 2)]:
tm.assert_index_equal(res, exp)
assert res.freq is None
index = pd.date_range('2001-01-01', periods=2, freq='2D', tz=tz)
exp = pd.DatetimeIndex(
['2001-01-01', '2001-01-01', '2001-01-03', '2001-01-03'], tz=tz)
for res in [index.repeat(2), np.repeat(index, 2)]:
tm.assert_index_equal(res, exp)
assert res.freq is None
index = pd.DatetimeIndex(['2001-01-01', 'NaT', '2003-01-01'], tz=tz)
exp = pd.DatetimeIndex([
'2001-01-01', '2001-01-01', '2001-01-01', 'NaT', 'NaT', 'NaT',
'2003-01-01', '2003-01-01', '2003-01-01'
],
tz=tz)
for res in [index.repeat(3), np.repeat(index, 3)]:
tm.assert_index_equal(res, exp)
assert res.freq is None
def test_repeat(self, tz_fixture):
tz = tz_fixture
reps = 2
msg = "the 'axis' parameter is not supported"
rng = pd.date_range(start='2016-01-01', periods=2, freq='30Min', tz=tz)
expected_rng = DatetimeIndex([
Timestamp('2016-01-01 00:00:00', tz=tz, freq='30T'),
Timestamp('2016-01-01 00:00:00', tz=tz, freq='30T'),
Timestamp('2016-01-01 00:30:00', tz=tz, freq='30T'),
Timestamp('2016-01-01 00:30:00', tz=tz, freq='30T'),
])
res = rng.repeat(reps)
tm.assert_index_equal(res, expected_rng)
assert res.freq is None
tm.assert_index_equal(np.repeat(rng, reps), expected_rng)
tm.assert_raises_regex(ValueError, msg, np.repeat, rng, reps, axis=1)
def test_resolution(self, tz_fixture):
tz = tz_fixture
for freq, expected in zip(
['A', 'Q', 'M', 'D', 'H', 'T', 'S', 'L', 'U'], [
'day', 'day', 'day', 'day', 'hour', 'minute', 'second',
'millisecond', 'microsecond'
]):
idx = pd.date_range(start='2013-04-01',
periods=30,
freq=freq,
tz=tz)
assert idx.resolution == expected
def test_value_counts_unique(self, tz_fixture):
tz = tz_fixture
# GH 7735
idx = pd.date_range('2011-01-01 09:00', freq='H', periods=10)
# create repeated values, 'n'th element is repeated by n+1 times
idx = DatetimeIndex(np.repeat(idx.values, range(1,
len(idx) + 1)),
tz=tz)
exp_idx = pd.date_range('2011-01-01 18:00',
freq='-1H',
periods=10,
tz=tz)
expected = Series(range(10, 0, -1), index=exp_idx, dtype='int64')
for obj in [idx, Series(idx)]:
tm.assert_series_equal(obj.value_counts(), expected)
expected = pd.date_range('2011-01-01 09:00',
freq='H',
periods=10,
tz=tz)
tm.assert_index_equal(idx.unique(), expected)
idx = DatetimeIndex([
'2013-01-01 09:00', '2013-01-01 09:00', '2013-01-01 09:00',
'2013-01-01 08:00', '2013-01-01 08:00', pd.NaT
],
tz=tz)
exp_idx = DatetimeIndex(['2013-01-01 09:00', '2013-01-01 08:00'],
tz=tz)
expected = Series([3, 2], index=exp_idx)
for obj in [idx, Series(idx)]:
tm.assert_series_equal(obj.value_counts(), expected)
exp_idx = DatetimeIndex(
['2013-01-01 09:00', '2013-01-01 08:00', pd.NaT], tz=tz)
expected = Series([3, 2, 1], index=exp_idx)
for obj in [idx, Series(idx)]:
tm.assert_series_equal(obj.value_counts(dropna=False), expected)
tm.assert_index_equal(idx.unique(), exp_idx)
def test_nonunique_contains(self):
# GH 9512
for idx in map(DatetimeIndex,
([0, 1, 0], [0, 0, -1], [0, -1, -1],
['2015', '2015', '2016'], ['2015', '2015', '2014'])):
assert idx[0] in idx
@pytest.mark.parametrize('idx', [
DatetimeIndex(
['2011-01-01', '2011-01-02', '2011-01-03'], freq='D', name='idx'),
DatetimeIndex(
['2011-01-01 09:00', '2011-01-01 10:00', '2011-01-01 11:00'],
freq='H',
name='tzidx',
tz='Asia/Tokyo')
])
def test_order_with_freq(self, idx):
ordered = idx.sort_values()
tm.assert_index_equal(ordered, idx)
assert ordered.freq == idx.freq
ordered = idx.sort_values(ascending=False)
expected = idx[::-1]
tm.assert_index_equal(ordered, expected)
assert ordered.freq == expected.freq
assert ordered.freq.n == -1
ordered, indexer = idx.sort_values(return_indexer=True)
tm.assert_index_equal(ordered, idx)
tm.assert_numpy_array_equal(indexer,
np.array([0, 1, 2]),
check_dtype=False)
assert ordered.freq == idx.freq
ordered, indexer = idx.sort_values(return_indexer=True,
ascending=False)
expected = idx[::-1]
tm.assert_index_equal(ordered, expected)
tm.assert_numpy_array_equal(indexer,
np.array([2, 1, 0]),
check_dtype=False)
assert ordered.freq == expected.freq
assert ordered.freq.n == -1
@pytest.mark.parametrize(
'index_dates,expected_dates',
[([
'2011-01-01', '2011-01-03', '2011-01-05', '2011-01-02',
'2011-01-01'
], [
'2011-01-01', '2011-01-01', '2011-01-02', '2011-01-03',
'2011-01-05'
]),
([
'2011-01-01', '2011-01-03', '2011-01-05', '2011-01-02',
'2011-01-01'
], [
'2011-01-01', '2011-01-01', '2011-01-02', '2011-01-03',
'2011-01-05'
]),
([pd.NaT, '2011-01-03', '2011-01-05', '2011-01-02', pd.NaT
], [pd.NaT, pd.NaT, '2011-01-02', '2011-01-03', '2011-01-05'])])
def test_order_without_freq(self, index_dates, expected_dates, tz_fixture):
tz = tz_fixture
# without freq
index = DatetimeIndex(index_dates, tz=tz, name='idx')
expected = DatetimeIndex(expected_dates, tz=tz, name='idx')
ordered = index.sort_values()
tm.assert_index_equal(ordered, expected)
assert ordered.freq is None
ordered = index.sort_values(ascending=False)
tm.assert_index_equal(ordered, expected[::-1])
assert ordered.freq is None
ordered, indexer = index.sort_values(return_indexer=True)
tm.assert_index_equal(ordered, expected)
exp = np.array([0, 4, 3, 1, 2])
tm.assert_numpy_array_equal(indexer, exp, check_dtype=False)
assert ordered.freq is None
ordered, indexer = index.sort_values(return_indexer=True,
ascending=False)
tm.assert_index_equal(ordered, expected[::-1])
exp = np.array([2, 1, 3, 4, 0])
tm.assert_numpy_array_equal(indexer, exp, check_dtype=False)
assert ordered.freq is None
def test_drop_duplicates_metadata(self):
# GH 10115
idx = pd.date_range('2011-01-01', '2011-01-31', freq='D', name='idx')
result = idx.drop_duplicates()
tm.assert_index_equal(idx, result)
assert idx.freq == result.freq
idx_dup = idx.append(idx)
assert idx_dup.freq is None # freq is reset
result = idx_dup.drop_duplicates()
tm.assert_index_equal(idx, result)
assert result.freq is None
def test_drop_duplicates(self):
# to check Index/Series compat
base = pd.date_range('2011-01-01', '2011-01-31', freq='D', name='idx')
idx = base.append(base[:5])
res = idx.drop_duplicates()
tm.assert_index_equal(res, base)
res = Series(idx).drop_duplicates()
tm.assert_series_equal(res, Series(base))
res = idx.drop_duplicates(keep='last')
exp = base[5:].append(base[:5])
tm.assert_index_equal(res, exp)
res = Series(idx).drop_duplicates(keep='last')
tm.assert_series_equal(res, Series(exp, index=np.arange(5, 36)))
res = idx.drop_duplicates(keep=False)
tm.assert_index_equal(res, base[5:])
res = Series(idx).drop_duplicates(keep=False)
tm.assert_series_equal(res, Series(base[5:], index=np.arange(5, 31)))
@pytest.mark.parametrize('freq', [
'A', '2A', '-2A', 'Q', '-1Q', 'M', '-1M', 'D', '3D', '-3D', 'W', '-1W',
'H', '2H', '-2H', 'T', '2T', 'S', '-3S'
])
def test_infer_freq(self, freq):
# GH 11018
idx = pd.date_range('2011-01-01 09:00:00', freq=freq, periods=10)
result = pd.DatetimeIndex(idx.asi8, freq='infer')
tm.assert_index_equal(idx, result)
assert result.freq == freq
def test_nat_new(self):
idx = pd.date_range('2011-01-01', freq='D', periods=5, name='x')
result = idx._nat_new()
exp = pd.DatetimeIndex([pd.NaT] * 5, name='x')
tm.assert_index_equal(result, exp)
result = idx._nat_new(box=False)
exp = np.array([tslib.iNaT] * 5, dtype=np.int64)
tm.assert_numpy_array_equal(result, exp)
def test_nat(self, tz_naive_fixture):
timezone = tz_naive_fixture
assert pd.DatetimeIndex._na_value is pd.NaT
assert pd.DatetimeIndex([])._na_value is pd.NaT
idx = pd.DatetimeIndex(['2011-01-01', '2011-01-02'], tz=timezone)
assert idx._can_hold_na
tm.assert_numpy_array_equal(idx._isnan, np.array([False, False]))
assert not idx.hasnans
tm.assert_numpy_array_equal(idx._nan_idxs, np.array([], dtype=np.intp))
idx = pd.DatetimeIndex(['2011-01-01', 'NaT'], tz=timezone)
assert idx._can_hold_na
tm.assert_numpy_array_equal(idx._isnan, np.array([False, True]))
assert idx.hasnans
tm.assert_numpy_array_equal(idx._nan_idxs, np.array([1],
dtype=np.intp))
def test_equals(self):
# GH 13107
idx = pd.DatetimeIndex(['2011-01-01', '2011-01-02', 'NaT'])
assert idx.equals(idx)
assert idx.equals(idx.copy())
assert idx.equals(idx.astype(object))
assert idx.astype(object).equals(idx)
assert idx.astype(object).equals(idx.astype(object))
assert not idx.equals(list(idx))
assert not idx.equals(pd.Series(idx))
idx2 = pd.DatetimeIndex(['2011-01-01', '2011-01-02', 'NaT'],
tz='US/Pacific')
assert not idx.equals(idx2)
assert not idx.equals(idx2.copy())
assert not idx.equals(idx2.astype(object))
assert not idx.astype(object).equals(idx2)
assert not idx.equals(list(idx2))
assert not idx.equals(pd.Series(idx2))
# same internal, different tz
idx3 = pd.DatetimeIndex._simple_new(idx.asi8, tz='US/Pacific')
tm.assert_numpy_array_equal(idx.asi8, idx3.asi8)
assert not idx.equals(idx3)
assert not idx.equals(idx3.copy())
assert not idx.equals(idx3.astype(object))
assert not idx.astype(object).equals(idx3)
assert not idx.equals(list(idx3))
assert not idx.equals(pd.Series(idx3))
@pytest.mark.parametrize('values',
[['20180101', '20180103', '20180105'], []])
@pytest.mark.parametrize(
'freq', ['2D', Day(2), '2B',
BDay(2), '48H', Hour(48)])
@pytest.mark.parametrize('tz', [None, 'US/Eastern'])
def test_freq_setter(self, values, freq, tz):
# GH 20678
idx = DatetimeIndex(values, tz=tz)
# can set to an offset, converting from string if necessary
idx.freq = freq
assert idx.freq == freq
assert isinstance(idx.freq, ABCDateOffset)
# can reset to None
idx.freq = None
assert idx.freq is None
def test_freq_setter_errors(self):
# GH 20678
idx = DatetimeIndex(['20180101', '20180103', '20180105'])
# setting with an incompatible freq
msg = ('Inferred frequency 2D from passed values does not conform to '
'passed frequency 5D')
with tm.assert_raises_regex(ValueError, msg):
idx.freq = '5D'
# setting with non-freq string
with tm.assert_raises_regex(ValueError, 'Invalid frequency'):
idx.freq = 'foo'
def test_offset_deprecated(self):
# GH 20716
idx = pd.DatetimeIndex(['20180101', '20180102'])
# getter deprecated
with tm.assert_produces_warning(FutureWarning):
idx.offset
# setter deprecated
with tm.assert_produces_warning(FutureWarning):
idx.offset = BDay()
#pandas sequence
pd.date_range('2019-07-01', '2019-10-30')
pd.date_range('2019-07-01', periods=45)
pd.date_range('2019-07-01', periods=3, freq='M')
pd.date_range('2019-07-01', periods=5, freq='H')
#%%%
#
#%%%
pd.timedelta_range(0, periods=9, freq='2H20T')
#business day offser
from pandas.tseries.offsets import BDay
pd.date_range('2019-07-01', periods=9, freq=BDay())
#see the gap in days - Sat & SUn
#%%%
#using Frequencies and Offsets
#%%%
#Reading Stock Data
#conda install pandas-datareader
from pandas_datareader import data
#https://pandas-datareader.readthedocs.io/en/latest/
#https://pandas-datareader.readthedocs.io/en/latest/remote_data.html#remote-data-google
import pandas as pd
from tia.bbg import LocalTerminal
import tia.bbg.datamgr as dm
import os
from pandas import ExcelWriter
import datetime as dt
from pandas.tseries.offsets import BDay
import glob
t1 = dt.date.today() - BDay(1)
t1 = t1.strftime('%Y%m%d')
mgr = dm.BbgDataManager(
) #this is used to access the bloomberg api with python, used in getAdvs method in class
class executedOrderReport(object):
def __init__(self,
location,
saveLoc,
threshold,
advThreshold,
delimiter="|",
*args,
**kwargs):
"""
takes 3 parameters
location = where the raw fidessa file is
saveLoc = where the output report will go
"""
self.location = location
self.delimiter = delimiter
def close(self, date):
"""Return stock close price for given date."""
d = date if is_bday(date) else (date - BDay(1)).date()
return self._prices.loc[d]["Close"]
class TestDatetimeIndexOps(Ops):
def setup_method(self, method):
super().setup_method(method)
mask = lambda x: (isinstance(x, DatetimeIndex) or isinstance(x, PeriodIndex))
self.is_valid_objs = [o for o in self.objs if mask(o)]
self.not_valid_objs = [o for o in self.objs if not mask(o)]
def test_ops_properties(self):
f = lambda x: isinstance(x, DatetimeIndex)
self.check_ops_properties(DatetimeIndex._field_ops, f)
self.check_ops_properties(DatetimeIndex._object_ops, f)
self.check_ops_properties(DatetimeIndex._bool_ops, f)
def test_ops_properties_basic(self):
# sanity check that the behavior didn't change
# GH#7206
for op in ["year", "day", "second", "weekday"]:
msg = f"'Series' object has no attribute '{op}'"
with pytest.raises(AttributeError, match=msg):
getattr(self.dt_series, op)
# attribute access should still work!
s = Series(dict(year=2000, month=1, day=10))
assert s.year == 2000
assert s.month == 1
assert s.day == 10
msg = "'Series' object has no attribute 'weekday'"
with pytest.raises(AttributeError, match=msg):
s.weekday
def test_repeat_range(self, tz_naive_fixture):
tz = tz_naive_fixture
rng = date_range("1/1/2000", "1/1/2001")
result = rng.repeat(5)
assert result.freq is None
assert len(result) == 5 * len(rng)
index = pd.date_range("2001-01-01", periods=2, freq="D", tz=tz)
exp = pd.DatetimeIndex(
["2001-01-01", "2001-01-01", "2001-01-02", "2001-01-02"], tz=tz
)
for res in [index.repeat(2), np.repeat(index, 2)]:
tm.assert_index_equal(res, exp)
assert res.freq is None
index = pd.date_range("2001-01-01", periods=2, freq="2D", tz=tz)
exp = pd.DatetimeIndex(
["2001-01-01", "2001-01-01", "2001-01-03", "2001-01-03"], tz=tz
)
for res in [index.repeat(2), np.repeat(index, 2)]:
tm.assert_index_equal(res, exp)
assert res.freq is None
index = pd.DatetimeIndex(["2001-01-01", "NaT", "2003-01-01"], tz=tz)
exp = pd.DatetimeIndex(
[
"2001-01-01",
"2001-01-01",
"2001-01-01",
"NaT",
"NaT",
"NaT",
"2003-01-01",
"2003-01-01",
"2003-01-01",
],
tz=tz,
)
for res in [index.repeat(3), np.repeat(index, 3)]:
tm.assert_index_equal(res, exp)
assert res.freq is None
def test_repeat(self, tz_naive_fixture):
tz = tz_naive_fixture
reps = 2
msg = "the 'axis' parameter is not supported"
rng = pd.date_range(start="2016-01-01", periods=2, freq="30Min", tz=tz)
expected_rng = DatetimeIndex(
[
Timestamp("2016-01-01 00:00:00", tz=tz, freq="30T"),
Timestamp("2016-01-01 00:00:00", tz=tz, freq="30T"),
Timestamp("2016-01-01 00:30:00", tz=tz, freq="30T"),
Timestamp("2016-01-01 00:30:00", tz=tz, freq="30T"),
]
)
res = rng.repeat(reps)
tm.assert_index_equal(res, expected_rng)
assert res.freq is None
tm.assert_index_equal(np.repeat(rng, reps), expected_rng)
with pytest.raises(ValueError, match=msg):
np.repeat(rng, reps, axis=1)
def test_resolution(self, tz_naive_fixture):
tz = tz_naive_fixture
for freq, expected in zip(
["A", "Q", "M", "D", "H", "T", "S", "L", "U"],
[
"day",
"day",
"day",
"day",
"hour",
"minute",
"second",
"millisecond",
"microsecond",
],
):
idx = pd.date_range(start="2013-04-01", periods=30, freq=freq, tz=tz)
assert idx.resolution == expected
def test_value_counts_unique(self, tz_naive_fixture):
tz = tz_naive_fixture
# GH 7735
idx = pd.date_range("2011-01-01 09:00", freq="H", periods=10)
# create repeated values, 'n'th element is repeated by n+1 times
idx = DatetimeIndex(np.repeat(idx.values, range(1, len(idx) + 1)), tz=tz)
exp_idx = pd.date_range("2011-01-01 18:00", freq="-1H", periods=10, tz=tz)
expected = Series(range(10, 0, -1), index=exp_idx, dtype="int64")
for obj in [idx, Series(idx)]:
tm.assert_series_equal(obj.value_counts(), expected)
expected = pd.date_range("2011-01-01 09:00", freq="H", periods=10, tz=tz)
tm.assert_index_equal(idx.unique(), expected)
idx = DatetimeIndex(
[
"2013-01-01 09:00",
"2013-01-01 09:00",
"2013-01-01 09:00",
"2013-01-01 08:00",
"2013-01-01 08:00",
pd.NaT,
],
tz=tz,
)
exp_idx = DatetimeIndex(["2013-01-01 09:00", "2013-01-01 08:00"], tz=tz)
expected = Series([3, 2], index=exp_idx)
for obj in [idx, Series(idx)]:
tm.assert_series_equal(obj.value_counts(), expected)
exp_idx = DatetimeIndex(["2013-01-01 09:00", "2013-01-01 08:00", pd.NaT], tz=tz)
expected = Series([3, 2, 1], index=exp_idx)
for obj in [idx, Series(idx)]:
tm.assert_series_equal(obj.value_counts(dropna=False), expected)
tm.assert_index_equal(idx.unique(), exp_idx)
def test_nonunique_contains(self):
# GH 9512
for idx in map(
DatetimeIndex,
(
[0, 1, 0],
[0, 0, -1],
[0, -1, -1],
["2015", "2015", "2016"],
["2015", "2015", "2014"],
),
):
assert idx[0] in idx
@pytest.mark.parametrize(
"idx",
[
DatetimeIndex(
["2011-01-01", "2011-01-02", "2011-01-03"], freq="D", name="idx"
),
DatetimeIndex(
["2011-01-01 09:00", "2011-01-01 10:00", "2011-01-01 11:00"],
freq="H",
name="tzidx",
tz="Asia/Tokyo",
),
],
)
def test_order_with_freq(self, idx):
ordered = idx.sort_values()
tm.assert_index_equal(ordered, idx)
assert ordered.freq == idx.freq
ordered = idx.sort_values(ascending=False)
expected = idx[::-1]
tm.assert_index_equal(ordered, expected)
assert ordered.freq == expected.freq
assert ordered.freq.n == -1
ordered, indexer = idx.sort_values(return_indexer=True)
tm.assert_index_equal(ordered, idx)
tm.assert_numpy_array_equal(indexer, np.array([0, 1, 2]), check_dtype=False)
assert ordered.freq == idx.freq
ordered, indexer = idx.sort_values(return_indexer=True, ascending=False)
expected = idx[::-1]
tm.assert_index_equal(ordered, expected)
tm.assert_numpy_array_equal(indexer, np.array([2, 1, 0]), check_dtype=False)
assert ordered.freq == expected.freq
assert ordered.freq.n == -1
@pytest.mark.parametrize(
"index_dates,expected_dates",
[
(
["2011-01-01", "2011-01-03", "2011-01-05", "2011-01-02", "2011-01-01"],
["2011-01-01", "2011-01-01", "2011-01-02", "2011-01-03", "2011-01-05"],
),
(
["2011-01-01", "2011-01-03", "2011-01-05", "2011-01-02", "2011-01-01"],
["2011-01-01", "2011-01-01", "2011-01-02", "2011-01-03", "2011-01-05"],
),
(
[pd.NaT, "2011-01-03", "2011-01-05", "2011-01-02", pd.NaT],
[pd.NaT, pd.NaT, "2011-01-02", "2011-01-03", "2011-01-05"],
),
],
)
def test_order_without_freq(self, index_dates, expected_dates, tz_naive_fixture):
tz = tz_naive_fixture
# without freq
index = DatetimeIndex(index_dates, tz=tz, name="idx")
expected = DatetimeIndex(expected_dates, tz=tz, name="idx")
ordered = index.sort_values()
tm.assert_index_equal(ordered, expected)
assert ordered.freq is None
ordered = index.sort_values(ascending=False)
tm.assert_index_equal(ordered, expected[::-1])
assert ordered.freq is None
ordered, indexer = index.sort_values(return_indexer=True)
tm.assert_index_equal(ordered, expected)
exp = np.array([0, 4, 3, 1, 2])
tm.assert_numpy_array_equal(indexer, exp, check_dtype=False)
assert ordered.freq is None
ordered, indexer = index.sort_values(return_indexer=True, ascending=False)
tm.assert_index_equal(ordered, expected[::-1])
exp = np.array([2, 1, 3, 4, 0])
tm.assert_numpy_array_equal(indexer, exp, check_dtype=False)
assert ordered.freq is None
def test_drop_duplicates_metadata(self):
# GH 10115
idx = pd.date_range("2011-01-01", "2011-01-31", freq="D", name="idx")
result = idx.drop_duplicates()
tm.assert_index_equal(idx, result)
assert idx.freq == result.freq
idx_dup = idx.append(idx)
assert idx_dup.freq is None # freq is reset
result = idx_dup.drop_duplicates()
tm.assert_index_equal(idx, result)
assert result.freq is None
def test_drop_duplicates(self):
# to check Index/Series compat
base = pd.date_range("2011-01-01", "2011-01-31", freq="D", name="idx")
idx = base.append(base[:5])
res = idx.drop_duplicates()
tm.assert_index_equal(res, base)
res = Series(idx).drop_duplicates()
tm.assert_series_equal(res, Series(base))
res = idx.drop_duplicates(keep="last")
exp = base[5:].append(base[:5])
tm.assert_index_equal(res, exp)
res = Series(idx).drop_duplicates(keep="last")
tm.assert_series_equal(res, Series(exp, index=np.arange(5, 36)))
res = idx.drop_duplicates(keep=False)
tm.assert_index_equal(res, base[5:])
res = Series(idx).drop_duplicates(keep=False)
tm.assert_series_equal(res, Series(base[5:], index=np.arange(5, 31)))
@pytest.mark.parametrize(
"freq",
[
"A",
"2A",
"-2A",
"Q",
"-1Q",
"M",
"-1M",
"D",
"3D",
"-3D",
"W",
"-1W",
"H",
"2H",
"-2H",
"T",
"2T",
"S",
"-3S",
],
)
def test_infer_freq(self, freq):
# GH 11018
idx = pd.date_range("2011-01-01 09:00:00", freq=freq, periods=10)
result = pd.DatetimeIndex(idx.asi8, freq="infer")
tm.assert_index_equal(idx, result)
assert result.freq == freq
def test_nat(self, tz_naive_fixture):
tz = tz_naive_fixture
assert pd.DatetimeIndex._na_value is pd.NaT
assert pd.DatetimeIndex([])._na_value is pd.NaT
idx = pd.DatetimeIndex(["2011-01-01", "2011-01-02"], tz=tz)
assert idx._can_hold_na
tm.assert_numpy_array_equal(idx._isnan, np.array([False, False]))
assert idx.hasnans is False
tm.assert_numpy_array_equal(idx._nan_idxs, np.array([], dtype=np.intp))
idx = pd.DatetimeIndex(["2011-01-01", "NaT"], tz=tz)
assert idx._can_hold_na
tm.assert_numpy_array_equal(idx._isnan, np.array([False, True]))
assert idx.hasnans is True
tm.assert_numpy_array_equal(idx._nan_idxs, np.array([1], dtype=np.intp))
def test_equals(self):
# GH 13107
idx = pd.DatetimeIndex(["2011-01-01", "2011-01-02", "NaT"])
assert idx.equals(idx)
assert idx.equals(idx.copy())
assert idx.equals(idx.astype(object))
assert idx.astype(object).equals(idx)
assert idx.astype(object).equals(idx.astype(object))
assert not idx.equals(list(idx))
assert not idx.equals(pd.Series(idx))
idx2 = pd.DatetimeIndex(["2011-01-01", "2011-01-02", "NaT"], tz="US/Pacific")
assert not idx.equals(idx2)
assert not idx.equals(idx2.copy())
assert not idx.equals(idx2.astype(object))
assert not idx.astype(object).equals(idx2)
assert not idx.equals(list(idx2))
assert not idx.equals(pd.Series(idx2))
# same internal, different tz
idx3 = pd.DatetimeIndex._simple_new(idx.asi8, tz="US/Pacific")
tm.assert_numpy_array_equal(idx.asi8, idx3.asi8)
assert not idx.equals(idx3)
assert not idx.equals(idx3.copy())
assert not idx.equals(idx3.astype(object))
assert not idx.astype(object).equals(idx3)
assert not idx.equals(list(idx3))
assert not idx.equals(pd.Series(idx3))
# check that we do not raise when comparing with OutOfBounds objects
oob = pd.Index([datetime(2500, 1, 1)] * 3, dtype=object)
assert not idx.equals(oob)
assert not idx2.equals(oob)
assert not idx3.equals(oob)
# check that we do not raise when comparing with OutOfBounds dt64
oob2 = oob.map(np.datetime64)
assert not idx.equals(oob2)
assert not idx2.equals(oob2)
assert not idx3.equals(oob2)
@pytest.mark.parametrize("values", [["20180101", "20180103", "20180105"], []])
@pytest.mark.parametrize("freq", ["2D", Day(2), "2B", BDay(2), "48H", Hour(48)])
@pytest.mark.parametrize("tz", [None, "US/Eastern"])
def test_freq_setter(self, values, freq, tz):
# GH 20678
idx = DatetimeIndex(values, tz=tz)
# can set to an offset, converting from string if necessary
idx._data.freq = freq
assert idx.freq == freq
assert isinstance(idx.freq, ABCDateOffset)
# can reset to None
idx._data.freq = None
assert idx.freq is None
def test_freq_setter_errors(self):
# GH 20678
idx = DatetimeIndex(["20180101", "20180103", "20180105"])
# setting with an incompatible freq
msg = (
"Inferred frequency 2D from passed values does not conform to "
"passed frequency 5D"
)
with pytest.raises(ValueError, match=msg):
idx._data.freq = "5D"
# setting with non-freq string
with pytest.raises(ValueError, match="Invalid frequency"):
idx._data.freq = "foo"
def GetBulkRiskiness(stocks, start, end):
import pandas as pd
start_ = pd.DataFrame(columns = ['PriceRisk', 'ReturnRisk', 'Volatility'])
for stock in stocks:
_ = GetRiskiness(stock, start, end)
start_ = pd.concat([start_,_])
return start_
import pandas as pd
_days_of_week = {
0:"Monday",1:"Tuesday",2:"Wednesday",3:"Thursday",4:"Friday"
}
end = pd.datetime.today() - BDay(1)
#risk = GetBulkRiskiness(_stocks, start = pd.datetime.today() - BDay(60), end = pd.datetime.today())
try:
risk = pd.read_excel("Risk.xlsx")
risk.to_excel("Risk.xlsx")
except Exception:
pass
import pandas as pd
end = pd.datetime.today() - BDay(1)
Today = (GetTSXReturns(end - BDay(1),end))
Today.columns = ["Monday"]
end = pd.datetime.today() - BDay(2)
Thursday = (GetTSXReturns(end - BDay(1),end))
Thursday.columns = ["Friday"]
return str(mult) + code
#----------------------------------------------------------------------
# Offset names ("time rules") and related functions
from pandas.tseries.offsets import (Day, BDay, Hour, Minute, Second, Milli,
Week, Micro, MonthEnd, MonthBegin,
BMonthBegin, BMonthEnd, YearBegin, YearEnd,
BYearBegin, BYearEnd, QuarterBegin,
QuarterEnd, BQuarterBegin, BQuarterEnd)
_offset_map = {
'D' : Day(),
'B' : BDay(),
'H' : Hour(),
'T' : Minute(),
'S' : Second(),
'L' : Milli(),
'U' : Micro(),
None : None,
# Monthly - Calendar
'M' : MonthEnd(),
'MS' : MonthBegin(),
# Monthly - Business
'BM' : BMonthEnd(),
'BMS' : BMonthBegin(),
dateFormater = ConciseDateFormatter(AutoDateLocator())
palette = sns.color_palette()
#%%
rki, meldedatum, hospital = read_case_data("berlin-cases.csv",
"berlin-cases-meldedatum.csv",
"berlin-hospital.csv")
#%% Activity participation
from pandas.tseries.offsets import BDay
isBusinessDay = BDay().onOffset
act = pd.read_csv(
"C:/home/Development/matsim-org/matsim-episim/output/BerlinSnzData_daily_until20200705.csv",
sep="\t",
parse_dates=[0])
act_week = act[act.date.map(isBusinessDay)]
act_wend = act[act.date.map(lambda *args: not isBusinessDay(*args))]
fig, ax = plt.subplots(dpi=250, figsize=(7.5, 3.8))
ax = sns.scatterplot(x="date",
y="home",
label="home",
s=40,
def test_freq_offsets():
off = BDay(1, offset=timedelta(0, 1800))
assert off.freqstr == "B+30Min"
off = BDay(1, offset=timedelta(0, -1800))
assert off.freqstr == "B-30Min"
def mainloop(trackit_api_username, jira_rest_call_post,
jira_rest_call_get_trackit_id, jira_authorization, db_connection,
jira_key, track_it_full_hostname, jira_server_address, sql,
attachment_folder, duedate_map):
"""
Purpose:
The main function to run everything above. From retrieving Track-It!
data from the database to POSTing it to JIRA and closing Track-It!
work orders with a comment to the new JIRA issue.
Args:
trackit_api_username (str): A Track-It! Technician ID
jira_rest_call_post (str): POST-able URL for JIRA
jira_rest_call_get_trackit_id (str): GET-able URL for JIRA
jira_authorization (str): BASE64 Encoded Username:Password
db_connection (pymssql Object): A very specific library object from
pymssql-2.1.3-cp36-cp36m-win_amd64.whl which can connect to SQL
Server 2008.
jira_key (str): JIRA Project key
track_it_full_hostname (str): Track-It! Server Address/URL
jira_server_address (str): JIRA Server Address/URL
sql (str): A very specific SQL Select statement.
Example:
use TRACKIT_DATA;
SELECT wo_num 'Workorder Number',
priority 'Priority',
LEFT(CONVERT(VARCHAR, REQDATE, 120), 10) AS 'Request Date',
task AS 'Summary',
request 'Requestor',
RESPONS AS 'Assignee Username',
LEFT(CONVERT(VARCHAR, duedate, 120), 10) AS 'Due Date',
LEFT(CONVERT(VARCHAR, modidate, 120), 10) AS 'Modify Date',
TRACKIT_DATA.dbo.tasks.dept,
type,
wotype2 'Subtype',
wotype3 'Category',
respons 'Assigned Technician',
descript 'Description',
note 'Notes',
lookup1 'Company'
FROM TRACKIT_DATA.dbo.tasks
WHERE tasks.respons in ('Maxim Tam')
and priority in ('Ongoing Support','High','Urgent','Critical','Routine','Project')
and WorkOrderStatusId = 1
and reqdate >= '2017-07-11'
ORDER BY RESPONS, WO_NUM DESC;
attachment_folder (str): Folder path to a specific work order's attachments
duedate_map (str): Dictionary of {Issue Priority:Resolution Days}
Returns:
Nothing.
"""
# Declare database query output
database_full_output = get_database_cursor(db_connection, sql, jira_key)
print(datetime.now().strftime("%Y-%m-%d %H:%M:%S") +
" Amount of open workorders: " + str(len(database_full_output)))
# Cleans query output to create list of new Trackit workorder IDs
# Retrieves all existing Jira Trackit IDs and compares the two lists
trackit_ids_trackit = [
int(x["fields"]["customfield_10411"]) for x in database_full_output
]
trackit_ids_jira_dict = dict((get_request(jira_rest_call_get_trackit_id,
jira_authorization)))
trackit_ids_jira = [
int(issue["fields"]["customfield_10411"])
for issue in trackit_ids_jira_dict["issues"]
if str(issue["fields"]["customfield_10411"]) != 'None'
]
invalid_ids = [x for x in trackit_ids_trackit if x in trackit_ids_jira]
database_full_output_valid = [
x for x in database_full_output
if int(x["fields"]["customfield_10411"]) not in invalid_ids
]
# Submits POST request to JIRA to create new issue
# and closes & comments on the old Track-It! ticket
if len(database_full_output_valid) > 0:
for data in database_full_output_valid:
print("Moving workorders to Jira:" +
str(data["fields"]["customfield_10411"]))
logging.info("Moving workorders to Jira:" +
str(data["fields"]["customfield_10411"]))
trackit_key = create_trackit_key(
trackit_api_username,
track_it_full_hostname=track_it_full_hostname)
try:
response = (post_request(jira_rest_call_post, data,
jira_authorization))
except HTTPError:
# replace customfield with data if you want the post request json string
logging.error(
str(sys.exc_info()[1]) + ": " +
str(data["fields"]["customfield_10411"]))
continue
# Jira URL to newly created ticket
response = (response)
jira_link = "http://" + jira_server_address + \
"/browse/" + str(response["key"])
jira_attachment_link = "http://" + jira_server_address + r"/rest/api/2/issue/" + str(
response["key"]) + "/attachments"
print(jira_attachment_link)
logging.info("Successfully migrated to Jira at: " +
jira_attachment_link)
import_attachments(data["fields"]["customfield_10411"],
jira_attachment_link, jira_authorization,
attachment_folder)
post_addnote_request_trackit(trackit_key,
data["fields"]["customfield_10411"],
jira_link, track_it_full_hostname)
post_close_request_trackit(trackit_key,
data["fields"]["customfield_10411"],
jira_link, track_it_full_hostname)
# Attempts to close TrackIt tickets when previously unable to
if len(invalid_ids) > 0:
print("Updating previously locked workorders: " + str(invalid_ids))
for keys in invalid_ids:
trackit_key = create_trackit_key(trackit_api_username,
track_it_full_hostname)
jira_trackit_id_url = "http://" + jira_server_address + \
"/rest/api/2/search?jql=%22TrackIT%20%23%22%3D" + \
str(keys)
jira_link = "http://" + jira_server_address + "/browse/" \
+ get_request(jira_trackit_id_url,
jira_authorization)["issues"][0]["key"]
post_addnote_request_trackit(trackit_key, keys, jira_link,
track_it_full_hostname)
post_close_request_trackit(trackit_key, keys, jira_link,
track_it_full_hostname)
# Due Date Creation
# @TODO: PLEASE REFACTOR TO SHRINK MAINLOOP
get_empty_duedates_url = "http://" + jira_server_address + \
"/rest/api/2/search?jql=project%20%3D%20" \
"" + jira_key + "" \
"%20AND%20duedate%20is%20EMPTY%20AND%20type%20%20%3D%20%22Incident%20Management%22"
duedates_response = get_request(get_empty_duedates_url, jira_authorization)
srq_ids_empty = [[
ticket["key"], ticket["fields"]["priority"]["name"],
pd.to_datetime(ticket["fields"]["created"][0:10])
] for ticket in duedates_response["issues"]]
for value in srq_ids_empty:
duedate = str(value[2] + BDay(duedate_map[str(value[1])]))[0:10]
headers = {
"Authorization": "Basic YXBpOlBhc3N3b3Jk",
"Content-Type": "application/json"
}
r = requests.put('http://' + config["jira_server_address"] + \
'/rest/api/2/issue/' + str(value[0]),
data=json.dumps({"fields": {"duedate": str(duedate)}}), headers=headers)
affirmation_list = ['Y', 'y', 'Yes', 'YES', 'yes']
negation_list = ['N', 'n', 'No', 'NO', 'no']
valid_answers = ['Y', 'y', 'Yes', 'YES', 'yes', 'N', 'n', 'No', 'NO', 'no']
# Explaining nature of script
print("------------------------------------------------")
print("This script will analyse a reverse MACD trading strategy "
"in a cryptocurrency portfolio")
print("The maximum absolute exposure per asset is 25%")
print("------------------------------------------------")
'''
###########
## DATES ##
###########
'''
previous_bday = pd.datetime.today() - BDay(1)
earliest_date = dt.datetime(2015, 8, 8) # Earliest date with data available
default_initial_date = dt.datetime(2016, 1, 1)
default_final_date = dt.datetime(2019, 4, 30)
initial_date, final_date = gdate(default_initial_date, default_final_date)
'''
####################
## GATHERING DATA ##
####################
'''
df_data = gdata()
df_data = df_data[initial_date:final_date]
df_data.drop(['nasdaq_close', 'nasdaq_return'], axis=1, inplace=True)
'''
########################
## GENERATING SIGNALS ##
prop = entity.property.add()
prop.name = name
prop.value.indexed = indx
prop.value.string_value = item
if __name__ == "__main__":
#launch example python DailyForecast.py
#Track time of the simulation
startTime = tt.time()
#First day of trading
nowTime = datetime.now(tz=timezone('US/Eastern')).time()
if nowTime >= time(19,00):
dayToPredict = datetime.now(tz=timezone('US/Eastern')) + BDay(1)
else:
dayToPredict = datetime.now(tz=timezone('US/Eastern')) + BDay(0)
print "\nPredicting %s\n"%dayToPredict.date()
logging.info("Predicting %s\n"%dayToPredict.date())
NPredPast = 10
history_len = 100 #days
saftey_days = 10
startOfPredictSim = dayToPredict - BDay(NPredPast)
endOfHistoricalDate = dayToPredict - BDay(1)
startOfHistoricalDate = startOfPredictSim - BDay(history_len+saftey_days)
def test_generate(self):
rng1 = list(generate_range(START, END, offset=BDay()))
rng2 = list(generate_range(START, END, time_rule='B'))
assert rng1 == rng2
def PredictPrices(prices: PricingData,
predictionMethod: int = 0,
daysForward: int = 5,
numberOfLearningPasses: int = 500):
#Simple procedure to test different prediction methods
assert (0 <= predictionMethod <= 2)
plot = PlotHelper()
if predictionMethod == 0: #Linear projection
print('Running Linear Projection model predicting ' +
str(daysForward) + ' days...')
modelDescription = prices.stockTicker + '_Linear_daysforward' + str(
daysForward)
predDF = prices.GetPriceHistory()
predDF['Average'] = (predDF['Open'] + predDF['High'] + predDF['Low'] +
predDF['Close']) / 4
d = predDF.index[-1]
for i in range(
0, daysForward
): #Add new days to the end for crystal ball predictions
predDF.loc[d + BDay(i + 1), 'Average_Predicted'] = 0
predDF['PastSlope'] = predDF['Average'].shift(
daysForward) / predDF['Average'].shift(daysForward * 2)
predDF['Average_Predicted'] = predDF['Average'].shift(
daysForward) * predDF['PastSlope']
predDF['PercentageDeviation'] = abs(
(predDF['Average'] - predDF['Average_Predicted']) /
predDF['Average'])
else:
SourceFieldList = ['High', 'Low', 'Open', 'Close']
if predictionMethod == 1: #LSTM learning
print('Running LSTM model predicting ' + str(daysForward) +
' days...')
SourceFieldList = None
UseLSTM = True
window_size = 10
modelDescription = prices.stockTicker + '_LSTM' + '_epochs' + str(
numberOfLearningPasses) + '_histwin' + str(
window_size) + '_daysforward' + str(daysForward)
elif predictionMethod == 2: #CNN Learning
print('Running CNN model predicting ' + str(daysForward) +
' days...')
UseLSTM = False
window_size = 16 * daysForward
modelDescription = prices.stockTicker + '_CNN' + '_epochs' + str(
numberOfLearningPasses) + '_histwin' + str(
window_size) + '_daysforward' + str(daysForward)
learningModule = StockPredictionNN(modelName=prices.stockTicker,
UseLSTM=UseLSTM)
learningModule.LoadSource(prices.GetPriceHistory(),
SourceFieldList=SourceFieldList,
window_size=window_size)
learningModule.LoadTarget(targetDF=None,
prediction_target_days=daysForward)
learningModule.MakeBatches(batch_size=32, train_test_split=.93)
learningModule.Train(epochs=numberOfLearningPasses)
learningModule.Predict(True)
predDF = learningModule.GetTrainingResults(True, True)
averageDeviation = predDF['PercentageDeviation'].tail(
round(predDF.shape[0] /
4)).mean() #Average of the last 25% to account for training.
print('Average deviation: ', averageDeviation * 100, '%')
predDF = predDF.reindex(sorted(predDF.columns),
axis=1) #Sort columns alphabetical
predDF.to_csv(dataFolder + modelDescription + '.csv')
plot.PlotDataFrame(predDF[['Average',
'Average_Predicted']], modelDescription, 'Date',
'Price', True, 'experiment/' + modelDescription)
plot.PlotDataFrameDateRange(predDF[['Average', 'Average_Predicted']], None,
160, modelDescription + '_last160ays', 'Date',
'Price',
dataFolder + modelDescription + '_last160Days')
plot.PlotDataFrameDateRange(
predDF[['Average', 'Average_Predicted']], None, 1000,
modelDescription + '_last1000ays', 'Date', 'Price',
dataFolder + modelDescription + '_last1000Days')
def get_expiries_from_dates(self, date_time_index, calendar, tenor):
freq = self.get_business_days_tenor(tenor)
return pandas.DatetimeIndex(date_time_index + BDay(freq))
def main():
demo = 'd286f23fd3d3c4fbd6cc5768c2a6388d'
#data = read_csv('/Users/alenshaju/Downloads/SP500_tickers_100.csv')
#companies = data['Ticker'].to_list()[:10]
consumer_companies = [
'TJX', 'NKE', 'TGT', 'HD', 'LOW', 'PG', 'WMT', 'COST', 'MDLZ', 'EL',
'KO', 'PEP', 'PM', 'MO', 'BKNG', 'MCD', 'SBUX'
]
energy_companies = ['NEE', 'XOM', 'CVX']
fig_companies = [
'BLK', 'AXP', 'V', 'MA', 'PYPL', 'FIS', 'JPM', 'BAC', 'WFC', 'USB',
'SPGI', 'MS', 'SCHW', 'GS', 'BRK.B', 'AMT'
] #C
healthcare_companies = [
'ABBV', 'AMGN', 'GILD', 'ABT', 'DHR', 'MDT', 'SYK', 'ISRG', 'CVS',
'CI', 'TMO', 'UNH', 'ANTM', 'JNJ', 'PFE', 'LLY', 'BMY'
]
industrials_companies = [
'BA', 'RTX', 'LMT', 'DE', 'UPS', 'TSLA', 'GM', 'CAT', 'HON', 'GE',
'MMM', 'LIN', 'UNP'
]
tech_companies = [
'ADBE', 'CRM', 'INTU', 'GOOG', 'GOOG.L', 'FB', 'AMZN', 'ACN', 'IBM',
'AMAT', 'LRCX', 'NVDA', 'INTC', 'AVGO', 'TXN', 'QCOM', 'MU', 'AMD',
'MSFT', 'ORCL', 'NOW', 'AAPL'
]
mt_companies = ['CMCS.A', 'CHTR', 'CSCO', 'VZ', 'T', 'DIS', 'NFLX']
companies = ['UAL']
past_call_dict = {}
yec = YahooEarningsCalendar()
for company in companies:
print("Ticker:", company)
past_calls_df = get_past_earnings_call(yec, company)
past_call_dict[company] = past_calls_df
df_returns_scores = pd.DataFrame(columns=['Return', 'Score'])
sia = SentimentIntensityAnalyzer()
d = {}
with open(
"/Users/alenshaju/Downloads/LoughranMcDonald_MasterDictionary_2018.txt"
) as f:
for line in f:
(key, val) = line.split()
d[key] = float(val)
sia.lexicon.update(d)
excel_df = pd.DataFrame(
columns=['Ticker', 'Quarter', 'Sentiment Score', 'Returns'])
for company in companies:
print("For company: ", company)
for i, row in past_call_dict[company].iterrows():
date = datetime.datetime.strptime(row['startdatetime'],
'%Y-%m-%dT%H:%M:%S.%fZ')
quarter = pd.Timestamp(date).quarter
year = date.year
if year <= datetime.datetime.now().year:
if year == datetime.datetime.now().year:
if quarter >= pd.Timestamp(
datetime.datetime.now()).quarter:
continue
transcript = requests.get(
f'https://financialmodelingprep.com/api/v3/earning_call_transcript/{company}?quarter={quarter}&year={year}&apikey={demo}'
).json()
if len(transcript) == 0:
continue
transcript = transcript[0]['content'].split('\n')
if not bool(len(pd.bdate_range(date, date))):
date = date - BDay(1)
if (date + BDay(1)) in get_trading_close_holidays(year):
end_date = date + BDay(1)
else:
end_date = date
stock = yf.download(company,
start=date,
end=end_date + BDay(1) +
datetime.timedelta(1),
progress=False)
price_change_rate = (stock['Adj Close'][1] /
stock['Adj Close'][0]) - 1
price_change_percent = price_change_rate * 100
sentiment_score = sia.polarity_scores(
transcript[0])['pos'] - sia.polarity_scores(
transcript[0])['neg']
print(transcript)
print('score: ', sia.polarity_scores(transcript[0]))
print("price change: ", price_change_rate)
df_returns_scores = df_returns_scores.append(
{
'Return': price_change_rate,
'Score': sentiment_score
},
ignore_index=True)
excel_df = excel_df.append(
{
'Ticker': company,
"Date": date,
'Quarter': quarter,
'Sentiment Score': sentiment_score,
'Returns': price_change_rate
},
ignore_index=True)
if i > 8: # 10years - 4 quarters
break
excel_df.to_excel("/Users/alenshaju/Downloads/mt_excel_file_v1.xlsx")
x = df_returns_scores.Score.values.reshape(-1, 1)
y = df_returns_scores.Return.values.reshape(-1, 1)
x_train, x_test, y_train, y_test = train_test_split(x,
y,
test_size=0.4,
random_state=42)
support_vector_reg_model = SVR(kernel='rbf', C=100, gamma=0.1, epsilon=.1)
support_vector_reg_model.fit(x_train, y_train)
y_pred = support_vector_reg_model.predict(x_test)
mse = mean_squared_error(y_test, y_pred)
rmse = np.sqrt(mse)
r2_data = r2_score(y_test, y_pred)
print("Root mean square error: ", rmse)
print("R^2 score: ", r2_data)
train_test_label = ['Training Data', 'Testing Data']
model_color = ['m', 'c', 'g']
fig, axes = plt.subplots(nrows=1, ncols=2, figsize=(15, 10), sharey=True)
###### Training Data ##########
axes[0].plot(x_test,
y_pred,
color=model_color[0],
lw=2,
label='{} model'.format(train_test_label[0]))
axes[0].scatter(x_train[np.setdiff1d(np.arange(len(x_train)),
support_vector_reg_model.support_)],
y_train[np.setdiff1d(np.arange(len(x_train)),
support_vector_reg_model.support_)],
facecolor="none",
edgecolor=model_color[0],
s=50,
label='Training data')
axes[0].legend(loc='upper center',
bbox_to_anchor=(0.5, 1.1),
ncol=1,
fancybox=True,
shadow=True)
####### Testing Data #########
axes[1].plot(x_test,
y_pred,
color=model_color[1],
lw=2,
label='{} model'.format(train_test_label[1]))
axes[1].scatter(x_test[np.setdiff1d(np.arange(len(x_test)),
support_vector_reg_model.support_)],
y_pred[np.setdiff1d(np.arange(len(x_test)),
support_vector_reg_model.support_)],
facecolor="none",
edgecolor=model_color[1],
s=50,
label='Testing data')
axes[1].legend(loc='upper center',
bbox_to_anchor=(0.5, 1.1),
ncol=1,
fancybox=True,
shadow=True)
fig.text(0.5, 0.04, 'data', ha='center', va='center')
fig.text(0.06,
0.5,
'target',
ha='center',
va='center',
rotation='vertical')
fig.suptitle("Support Vector Regression", fontsize=14)
plt.show()
