def get_surprise_against_intraday_moves_over_event(
self,
data_frame_cross_orig,
cross,
event_fx,
event_name,
start,
end,
offset_list=[1, 5, 30, 60],
add_surprise=False,
surprise_field='survey-average'):
tsf = TimeSeriesFilter()
fields = [
'actual-release', 'survey-median', 'survey-average', 'survey-high',
'survey-low'
]
ef_time_frame = self.get_economic_event_date_time_fields(
fields, event_fx, event_name)
ef_time_frame = tsf.filter_time_series_by_date(start, end,
ef_time_frame)
return self.get_surprise_against_intraday_moves_over_custom_event(
data_frame_cross_orig,
ef_time_frame,
cross,
event_fx,
event_name,
start,
end,
offset_list=offset_list,
add_surprise=add_surprise,
surprise_field=surprise_field)
def get_time_series_cached(self, time_series_request):
"""
get_time_series_cached - Loads time series from cache (if it exists)
Parameters
----------
time_series_request : TimeSeriesRequest
contains various properties describing time series to fetched, including ticker, start & finish date etc.
Returns
-------
pandas.DataFrame
"""
if (time_series_request.freq == "intraday"):
ticker = time_series_request.tickers
else:
ticker = None
fname = self.create_time_series_hash_key(time_series_request, ticker)
if (fname in self._time_series_cache):
data_frame = self._time_series_cache[fname]
tsf = TimeSeriesFilter()
return tsf.filter_time_series(time_series_request, data_frame)
return None
def get_intraday_moves_over_event(self,
data_frame_rets,
cross,
event_fx,
event_name,
start,
end,
vol,
mins=3 * 60,
min_offset=0,
create_index=False,
resample=False,
freq='minutes'):
tsf = TimeSeriesFilter()
ef_time_frame = self.get_economic_event_date_time_dataframe(
event_fx, event_name)
ef_time_frame = tsf.filter_time_series_by_date(start, end,
ef_time_frame)
return self.get_intraday_moves_over_custom_event(
data_frame_rets,
ef_time_frame,
vol,
mins=mins,
min_offset=min_offset,
create_index=create_index,
resample=resample,
freq=freq) #, start, end)
def harvest_time_series(self, time_series_request, kill_session = True):
"""
havest_time_series - Loads time series from specified data provider
Parameters
----------
time_series_request : TimeSeriesRequest
contains various properties describing time series to fetched, including ticker, start & finish date etc.
Returns
-------
DataFrame
"""
tickers = time_series_request.tickers
loader = self.get_loader(time_series_request.data_source)
# check if tickers have been specified (if not load all of them for a category)
# also handle single tickers/list tickers
create_tickers = False
if tickers is None : create_tickers = True
elif isinstance(tickers, str):
if tickers == '': create_tickers = True
elif isinstance(tickers, list):
if tickers == []: create_tickers = True
if create_tickers:
time_series_request.tickers = self.config.get_tickers_list_for_category(
time_series_request.category, time_series_request.source, time_series_request.freq, time_series_request.cut)
# intraday or tick: only one ticker per cache file
if (time_series_request.freq in ['intraday', 'tick']):
data_frame_agg = self.download_intraday_tick(time_series_request, loader)
# daily: multiple tickers per cache file - assume we make one API call to vendor library
else: data_frame_agg = self.download_daily(time_series_request, loader)
if('internet_load' in time_series_request.cache_algo):
self.logger.debug("Internet loading.. ")
# signal to loader template to exit session
if loader is not None and kill_session == True: loader.kill_session()
if(time_series_request.cache_algo == 'cache_algo'):
self.logger.debug("Only caching data in memory, do not return any time series."); return
tsf = TimeSeriesFilter()
# only return time series if specified in the algo
if 'return' in time_series_request.cache_algo:
# special case for events/events-dt which is not indexed like other tables
if hasattr(time_series_request, 'category'):
if 'events' in time_series_request.category:
return data_frame_agg
try:
return tsf.filter_time_series(time_series_request, data_frame_agg)
except:
return None
def __init__(self):
# self.config = ConfigManager()
self.logger = LoggerManager().getLogger(__name__)
self.time_series_filter = TimeSeriesFilter()
self.time_series_io = TimeSeriesIO()
self._bbg_default_api = Constants().bbg_default_api
self._intraday_code = -1
return
def compare_strategy_vs_benchmark(self, br, strategy_df, benchmark_df):
"""
compare_strategy_vs_benchmark - Compares the trading strategy we are backtesting against a benchmark
Parameters
----------
br : BacktestRequest
Parameters for backtest such as start and finish dates
strategy_df : pandas.DataFrame
Strategy time series
benchmark_df : pandas.DataFrame
Benchmark time series
"""
include_benchmark = False
calc_stats = False
if hasattr(br, 'include_benchmark'): include_benchmark = br.include_benchmark
if hasattr(br, 'calc_stats'): calc_stats = br.calc_stats
if include_benchmark:
tsd = TimeSeriesDesc()
cash_backtest = CashBacktest()
ts_filter = TimeSeriesFilter()
ts_calcs = TimeSeriesCalcs()
# align strategy time series with that of benchmark
strategy_df, benchmark_df = strategy_df.align(benchmark_df, join='left', axis = 0)
# if necessary apply vol target to benchmark (to make it comparable with strategy)
if hasattr(br, 'portfolio_vol_adjust'):
if br.portfolio_vol_adjust is True:
benchmark_df = cash_backtest.calculate_vol_adjusted_index_from_prices(benchmark_df, br = br)
# only calculate return statistics if this has been specified (note when different frequencies of data
# might underrepresent vol
if calc_stats:
benchmark_df = benchmark_df.fillna(method='ffill')
tsd.calculate_ret_stats_from_prices(benchmark_df, br.ann_factor)
benchmark_df.columns = tsd.summary()
# realign strategy & benchmark
strategy_benchmark_df = strategy_df.join(benchmark_df, how='inner')
strategy_benchmark_df = strategy_benchmark_df.fillna(method='ffill')
strategy_benchmark_df = ts_filter.filter_time_series_by_date(br.plot_start, br.finish_date, strategy_benchmark_df)
strategy_benchmark_df = ts_calcs.create_mult_index_from_prices(strategy_benchmark_df)
self._benchmark_pnl = benchmark_df
self._benchmark_tsd = tsd
return strategy_benchmark_df
return strategy_df
def __init__(self):
super(EventStudy, self).__init__()
self.config = ConfigManager()
self.logger = LoggerManager().getLogger(__name__)
self.time_series_filter = TimeSeriesFilter()
self.time_series_io = TimeSeriesIO()
if (LightEventsFactory._econ_data_frame is None):
self.load_economic_events()
return
def get_economic_event_ret_over_custom_event_day(self,
data_frame_in,
event_dates,
name,
event,
start,
end,
lagged=False,
NYC_cutoff=10):
time_series_filter = TimeSeriesFilter()
event_dates = time_series_filter.filter_time_series_by_date(
start, end, event_dates)
data_frame = data_frame_in.copy(
deep=True) # because we change the dates!
time_series_tz = TimeSeriesTimezone()
calendar = Calendar()
bday = CustomBusinessDay(weekmask='Mon Tue Wed Thu Fri')
event_dates_nyc = time_series_tz.convert_index_from_UTC_to_new_york_time(
event_dates)
average_hour_nyc = numpy.average(event_dates_nyc.index.hour)
event_dates = calendar.floor_date(event_dates)
# realised is traditionally on later day eg. 3rd Jan realised ON is 2nd-3rd Jan realised
# so if Fed meeting is on 2nd Jan later, then we need realised labelled on 3rd (so minus a day)
# implied expires on next day eg. 3rd Jan implied ON is 3rd-4th Jan implied
# TODO smarter way of adjusting dates, as sometimes events can be before/after 10am NY cut
if (lagged and average_hour_nyc >= NYC_cutoff):
data_frame.index = data_frame.index - bday
elif (not lagged and average_hour_nyc < NYC_cutoff): # ie. implied
data_frame.index = data_frame.index + bday
# set as New York time and select only those ON vols at the 10am NY cut just before the event
data_frame_events = data_frame.ix[event_dates.index]
data_frame_events.columns = data_frame.columns.values + '-' + name + ' ' + event
return data_frame_events
def bus_day_of_month_seasonality(
self,
data_frame,
month_list=[1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12],
cum=True,
cal="FX",
partition_by_month=True,
add_average=False,
price_index=False):
tsc = TimeSeriesCalcs()
tsf = TimeSeriesFilter()
if price_index:
data_frame = data_frame.resample(
'B') # resample into business days
data_frame = tsc.calculate_returns(data_frame)
data_frame.index = pandas.to_datetime(data_frame.index)
data_frame = tsf.filter_time_series_by_holidays(data_frame, cal)
monthly_seasonality = tsc.average_by_month_day_by_bus_day(
data_frame, cal)
monthly_seasonality = monthly_seasonality.loc[month_list]
if partition_by_month:
monthly_seasonality = monthly_seasonality.unstack(level=0)
if add_average:
monthly_seasonality['Avg'] = monthly_seasonality.mean(axis=1)
if cum is True:
if partition_by_month:
monthly_seasonality.loc[0] = numpy.zeros(
len(monthly_seasonality.columns))
# monthly_seasonality.index = monthly_seasonality.index + 1 # shifting index
monthly_seasonality = monthly_seasonality.sort()
monthly_seasonality = tsc.create_mult_index(monthly_seasonality)
return monthly_seasonality
def dump_indicators(self):
tsf = TimeSeriesFilter()
self.logger.info("About to write all indicators to CSV")
self.indicator.to_csv(self._csv_indicator_dump, date_format='%d/%m/%Y')
if (self._csv_econ_indicator_dump is not None):
self.logger.info("About to write economy based indicators to CSV")
self.indicator_econ.to_csv(self._csv_econ_indicator_dump,
date_format='%d/%m/%Y')
self.logger.info("About to write final indicators to CSV")
# remove weekends and remove start of series
if (self._csv_final_indicator_dump is not None):
indicator_final_copy = tsf.filter_time_series_by_holidays(
self.indicator_final, cal='WEEKDAY')
indicator_final_copy = tsf.filter_time_series_by_date(
start_date="01 Jan 2000",
finish_date=None,
data_frame=indicator_final_copy)
indicator_final_copy.to_csv(self._csv_final_indicator_dump,
date_format='%d/%m/%Y')
groupby([date_index.month, date_index.day]).mean()
def group_by_year(self, data_frame):
date_index = data_frame.index
return data_frame.\
groupby([date_index.year])
def average_by_day_hour_min_by_bus_day(self, data_frame):
date_index = data_frame.index
return data_frame.\
groupby([Calendar().get_bus_day_of_month(date_index),
date_index.hour, date_index.minute]).mean()
def remove_NaN_rows(self, data_frame):
return data_frame.dropna()
if __name__ == '__main__':
tsc = TimeSeriesCalcs()
tsf = TimeSeriesFilter()
# test rolling ewma
date_range = pandas.bdate_range('2014-01-01', '2014-02-28')
print(tsc.get_bus_day_of_month(date_range))
foo = pandas.DataFrame(numpy.arange(0.0,13.0))
print(tsc.rolling_ewma(foo, span=3))
# fetch NFP times from Bloomberg
time_series_request = TimeSeriesRequest(
start_date = start_date, # start date
finish_date = finish_date, # finish date
category = "events",
freq = 'daily', # daily data
data_source = 'bloomberg', # use Bloomberg as data source
tickers = ['NFP'],
fields = ['release-date-time-full'], # which fields to download
vendor_tickers = ['NFP TCH Index'], # ticker (Bloomberg)
vendor_fields = ['ECO_FUTURE_RELEASE_DATE_LIST'], # which Bloomberg fields to download
cache_algo = 'internet_load_return') # how to return data
ltsf = LightTimeSeriesFactory()
ts_filter = TimeSeriesFilter()
df_event_times = ltsf.harvest_time_series(time_series_request)
utc_time = pytz.utc
df_event_times = pandas.DataFrame(index = df_event_times['NFP.release-date-time-full'])
df_event_times.index = df_event_times.index.tz_localize(utc_time) # work in UTC time
df_event_times = ts_filter.filter_time_series_by_date(start_date, finish_date, df_event_times)
# get last NFP time
start_date = df_event_times.index[-1] - timedelta(minutes=1)
finish_date = start_date + timedelta(minutes=4)
tickers = ['EURUSD', 'JPYUSD', 'GBPUSD']
vendor_tickers = ['EURUSD BGN Curncy', 'USDJPY BGN Curncy', 'GBPUSD BGN Curncy']
def get_intraday_moves_over_custom_event(self,
data_frame_rets,
ef_time_frame,
vol=False,
minute_start=5,
mins=3 * 60,
min_offset=0,
create_index=False,
resample=False,
freq='minutes'):
tsf = TimeSeriesFilter()
ef_time_frame = tsf.filter_time_series_by_date(
data_frame_rets.index[0], data_frame_rets.index[-1], ef_time_frame)
ef_time = ef_time_frame.index
if freq == 'minutes':
ef_time_start = ef_time - timedelta(minutes=minute_start)
ef_time_end = ef_time + timedelta(minutes=mins)
ann_factor = 252 * 1440
elif freq == 'days':
ef_time = ef_time_frame.index.normalize()
ef_time_start = ef_time - timedelta(days=minute_start)
ef_time_end = ef_time + timedelta(days=mins)
ann_factor = 252
ords = range(-minute_start + min_offset, mins + min_offset)
# all data needs to be equally spaced
if resample:
tsf = TimeSeriesFilter()
# make sure time series is properly sampled at 1 min intervals
data_frame_rets = data_frame_rets.resample('1min')
data_frame_rets = data_frame_rets.fillna(value=0)
data_frame_rets = tsf.remove_out_FX_out_of_hours(data_frame_rets)
data_frame_rets['Ind'] = numpy.nan
start_index = data_frame_rets.index.searchsorted(ef_time_start)
finish_index = data_frame_rets.index.searchsorted(ef_time_end)
# not all observation windows will be same length (eg. last one?)
# fill the indices which represent minutes
# TODO vectorise this!
for i in range(0, len(ef_time_frame.index)):
try:
data_frame_rets.ix[start_index[i]:finish_index[i],
'Ind'] = ords
except:
data_frame_rets.ix[start_index[i]:finish_index[i],
'Ind'] = ords[0:(finish_index[i] -
start_index[i])]
# set the release dates
data_frame_rets.ix[start_index, 'Rel'] = ef_time # set entry points
data_frame_rets.ix[finish_index + 1, 'Rel'] = numpy.zeros(
len(start_index)) # set exit points
data_frame_rets['Rel'] = data_frame_rets['Rel'].fillna(
method='pad') # fill down signals
data_frame_rets = data_frame_rets[pandas.notnull(
data_frame_rets['Ind'])] # get rid of other
data_frame = data_frame_rets.pivot(index='Ind',
columns='Rel',
values=data_frame_rets.columns[0])
data_frame.index.names = [None]
if create_index:
tsc = TimeSeriesCalcs()
data_frame.ix[-minute_start + min_offset, :] = numpy.nan
data_frame = tsc.create_mult_index(data_frame)
else:
if vol is True:
# annualise (if vol)
data_frame = pandas.rolling_std(
data_frame, window=5) * math.sqrt(ann_factor)
else:
data_frame = data_frame.cumsum()
return data_frame
def calculate_leverage_factor(self, returns_df, vol_target, vol_max_leverage, vol_periods = 60, vol_obs_in_year = 252,
vol_rebalance_freq = 'BM', data_resample_freq = None, data_resample_type = 'mean',
returns = True, period_shift = 0):
"""
calculate_leverage_factor - Calculates the time series of leverage for a specified vol target
Parameters
----------
returns_df : DataFrame
Asset returns
vol_target : float
vol target for assets
vol_max_leverage : float
maximum leverage allowed
vol_periods : int
number of periods to calculate volatility
vol_obs_in_year : int
number of observations in the year
vol_rebalance_freq : str
how often to rebalance
vol_resample_type : str
do we need to resample the underlying data first? (eg. have we got intraday data?)
returns : boolean
is this returns time series or prices?
period_shift : int
should we delay the signal by a number of periods?
Returns
-------
pandas.Dataframe
"""
tsc = TimeSeriesCalcs()
tsf = TimeSeriesFilter()
if data_resample_freq is not None:
return
# TODO not implemented yet
if not returns: returns_df = tsc.calculate_returns(returns_df)
roll_vol_df = tsc.rolling_volatility(returns_df,
periods = vol_periods, obs_in_year = vol_obs_in_year).shift(period_shift)
# calculate the leverage as function of vol target (with max lev constraint)
lev_df = vol_target / roll_vol_df
lev_df[lev_df > vol_max_leverage] = vol_max_leverage
lev_df = tsf.resample_time_series_frequency(lev_df, vol_rebalance_freq, data_resample_type)
returns_df, lev_df = returns_df.align(lev_df, join='left', axis = 0)
lev_df = lev_df.fillna(method='ffill')
lev_df.ix[0:vol_periods] = numpy.nan # ignore the first elements before the vol window kicks in
return lev_df
