DeltaHedgingStrategy#

class sigtech.framework.strategies.delta_hedging_strategy.DeltaHedgingStrategy

Baseclasses: DailyStrategy

A Delta hedging strategy. This strategy wraps an option strategy inheriting from OptionsStrategyBase and performs trading of a hedging instrument to match the delta for options from one group.

See also

sigtech.framework.strategies.single_options_strategy_base.OptionsStrategyBase.

The delta hedging parameters used are:

option_strategy_name: Name for the option strategy to match delta for.

option_group_name: Name of group to hedge delta for. If left blank and the option strategy has one group, this is taken from the strategy.

include_option_strategy: Flag to indicate if a position should be included in the option strategy.

delta_multiplier: Multiplier of delta to trade, -1 (i.e. 100% hedged) by default.

hedging_instrument_type: Instrument type to hedge with, the mapping is given below. Accepted value are 'FXSPOT', 'ETF', 'FUTURE', 'UNDERLYING', 'UNDERLYING_SPOT_START', or 'ROLLING_FX_FWD'. If 'UNDERLYING' is chosen, the hedging instrument is the option underlying (for options on futures).

hedging_frequency: Frequency of delta hedging, default is '1BD'.

delta_threshold: Minimal amount of required position update required to trigger placing a trade. Set to 0 by default.

rebalance_hedge_on_roll: Ignore the hedging_frequency and only rebalance the hedge on roll days.

custom_hedge_dates: Ignore the hedging_frequency and only rebalance the hedge this list of dates. This list is not adjusted in any way so please ensure a date is included to put on the initial hedge.

fractional_delta_threshold: If True, then the delta_threshold value is considered as a fractional deviation, else the absolute value of the difference in the delta values are compared (default is False).

max_absolute_delta: Maximal allowed hedged delta at any moment in time. Not used by default, and currently supported only for non-fx related hedges. If hedging instrument is 'UNDERLYING', then the limit is per underlying

size_at_execution: Flag used to specify if delta decision, sizing, and execution should happen within a very short period of time. This serves as a proxy to a real time data, when we have access only to one data point a day, and results in today rather than yesterday delta being hedged. Default is False, as the sizing, decision, and execution times are set around the group valuation time, and might happen at or after the market close, which could be a problem for a production strategy.

The following are the instruments used for delta hedging (if enabled), unless instrument type is 'FXSPOT' (OPTION_DELTA_HEDGING_INSTRUMENTS dictionary):

Underlying	Instrument Type	Instrument Name
`SPX INDEX`	`ETF`	`SPY UP EQUITY STRATEGY`
`SPX INDEX`	`FUTURE`	`USD ES INDEX LONG FRONT RF STRATEGY`
`NKY INDEX`	`FUTURE`	`JPY NK INDEX LONG FRONT RF STRATEGY`
`SX5E INDEX`	`FUTURE`	`EUR VG INDEX LONG FRONT RF STRATEGY`

To use another hedging underlying, or key, the DeltaHedgingStrategy.OPTION_DELTA_HEDGING_INSTRUMENTS dictionary should be updated:

import sigtech.framework as sig

sig.DeltaHedgingStrategy.OPTION_DELTA_HEDGING_INSTRUMENTS['NEW INDEX']={
    'FUTURE': 'NEW ROLLING FUTURE STRATEGY',
}

A rolling forward strategy can be used to hedge fx exposure. This is enabled by creating a strategy and updating the dictionary. For example:

import datetime as dtm
import sigtech.framework as sig

rolling_fwd = sig.RollingFXForwardStrategy(
    currency='USD',
    long_currency='EUR',
    forward_tenor='3M_IMM',
    start_date=dtm.date(2010, 1, 4)
)

sig.DeltaHedgingStrategy.OPTION_DELTA_HEDGING_INSTRUMENTS['EURUSD CURNCY'] = {'ROLLING_FX_FWD': rolling_fwd}

Example usage:

import datetime as dtm
import sigtech.framework as sig

start_date = dtm.date(2019, 1, 4)
end_date = dtm.date(2020, 12, 31)
group = sig.obj.get('SPX INDEX OTC OPTION GROUP')

rc = sig.RollingOptionStrategy(
    currency=group.underlying_obj.currency,
    start_date=start_date,
    end_date=end_date,
    group_name=group.name,
    option_type='Call',
    maturity='3M',
    rolling_frequencies=['1M'],
    strike_type='SPOT',
    strike='SPOT+10%',
    target_quantity=1
)

dhs = sig.DeltaHedgingStrategy(
    start_date=start_date,
    end_date=end_date,
    option_strategy_name=rc.name,
    delta_multiplier=-1,
    include_option_strategy=True,
    hedging_instrument_type='FUTURE',
    hedging_frequency='1BD',
)

currency: Optional[str]

custom_hedge_dates: Optional[list[datetime.date]]

delta_multiplier: Optional[float]

delta_threshold: Optional[float]

fractional_delta_threshold: Optional[bool]

hedging_frequency: Optional[str]

hedging_instrument_type: Literal['FXSPOT', 'FUTURE', 'ETF', 'UNDERLYING', 'ROLLING_FX_FWD', 'UNDERLYING_SPOT_START']

include_option_strategy: Optional[bool]

intraday_end_time: Optional[time]

intraday_start_time: Optional[time]

intraday_timezone: Optional[str]

max_absolute_delta: Optional[float]

OPTION_DELTA_HEDGING_INSTRUMENTS = {'NKY INDEX': {'FUTURE': <function nk_index_front>}, 'SPX INDEX': {'ETF': <function spy_equity_rs>, 'FUTURE': <function es_index_front>}, 'SX5E INDEX': {'FUTURE': <function vg_index_front>}}

option_group_name: Optional[str]

option_strategy: Optional[Strategy]

option_strategy_name: Optional[str]

rebalance_hedge_on_roll: Optional[bool]

size_at_execution: Optional[bool]

AVAILABLE_REBALANCE_FREQS

Subclasses: ScheduleFrequenciesWithIntraday

alias of ScheduleFrequencies

classmethod get_hedging_instrument(underlying_nameinstrument_type)

Return the hedging instrument.

Parameters:

underlying_name – Name of the underlying.
instrument_type – Type of instrument, e.g. 'FUTURE'.

greeks(fields=Nonedates=Nonedollar_greeks: Optional[bool] = Falseinclude_future_positions: Optional[bool] = Falsestart_date: Optional[date] = Noneend_date: Optional[date] = Nonedata_points: Optional[list] = None)

Information on the greeks for the options held in the strategy.

Parameters:

fields – List of greeks fields to output. If set to None, then Delta, Gamma and Vega are returned.
dates – dates/datetimes for which the greeks are computed. By default, the whole history dates at portfolio value time (23:59) are used.
dollar_greeks – provide dollar greeks, defaults to False.
include_future_positions – Boolean flag to include positions not yet traded (to get rough estimate of how much greeks will be needed in total to hedge overall position, once traded), defaults to False.
start_date – Start date - used if later than default start date, only if dates are not passed (optional).
end_date – End date - used if earlier than default end date, only if dates are not passed (optional).
data_points – Use data points induced valuations for intraday greeks. intraday is set to True internally, if data_points not empty (optional).

Returns:

pandas DataFrame.

hedge_option_delta(dt)

Hedge option delta.

Parameters:: dt – Reference datetime.
Returns:: Delta.

hedging_schedule()

Hedging schedule information.

metrics(fields=Nonedates=Nonedollar_greeks: Optional[bool] = Falseinclude_future_positions: Optional[bool] = Falsestart_date: Optional[date] = Noneend_date: Optional[date] = Nonedata_points: Optional[list] = None)

Collection of risk-based metrics for the options held in the strategy.

Parameters:

fields – List of fields to output. The default list is ['NPV', 'Delta', 'Gamma', 'Vega', 'ImpliedVolatility'].
dates – dates/datetimes for which the greeks are computed. By default, the whole history dates at portfolio value time (23:59) are used.
dollar_greeks – provide dollar greeks, defaults to False.
include_future_positions – Boolean flag to include positions not yet traded (to get rough estimate of how much greeks will be needed in total to hedge overall position, once traded), defaults to False.
start_date – Start date - used if later than default start date, only if dates are not passed (optional).
end_date – End date - used if earlier than default end date, only if dates are not passed (optional).
data_points – Use data points induced valuations for intraday greeks. intraday is set to True internally, if data_points not empty (optional).

Returns:

DataFrame of requested greeks (or NPV/ImpliedVolatility) per each day. The greeks are aggregated per underlying groups (second level column indexing is by group name). The NPV is for the whole strategy (second level index for NPV is the strategy name).

pnl_explain(fields=Nonedata_point=Nonestart_date=Noneend_date=None)

A breakdown of the actual pnl and greeks pnl over time for a list of fields.

Parameters:

fields – Input fields (optional).
data_point – Input data point (optional).
start_date – Start date - used if later than default start date (optional).
end_date – End date - used if earlier than default end date (optional).

Returns:

pandas DataFrame.

portfolio_greeks(fields=None, dates=None, dollar_greeks: Optional[bool] = False, include_future_positions: Optional[bool] = False, start_date: Optional[date] = None, end_date: Optional[date] = None, data_points: Optional[list] = None) deprecated

Deprecated method returning information on the greeks for the instruments held in the strategy.

schedule_information()

Schedule information for the strategy.

size_dt_from_decision_dt(decision_dt)

Return the previous strategy data point to the valuation date where the decision datetime falls in.

Parameters:: decision_dt – Decision datetime.
Returns:: Datetime.

strategy_initialization(dt)

Initial decision run on the start date of the strategy.

Parameters:: dt – Reference datetime.

trade_delta(dt)

Set a position in the hedging instrument to match the delta. delta_threshold refers to fractional deviation if fractional_delta_threshold is True and the simple difference between hedged and underlying delta otherwise.

Parameters:: dt – Reference datetime.
Returns:: Delta.

trade_delta_per_underlying(dt)

Parameters:: dt – Reference datetime.
Returns:: Delta.

trade_delta_per_underlying_spot_start(dt)

Parameters:: dt – Reference datetime.
Returns:: Delta.

trade_fx_delta(dt)

Perform an fx spot conversion to match the delta.

Parameters:: dt – Reference datetime.
Returns:: Delta.

trade_fx_delta_with_rolling_fwd(dt)

Set a position in the rolling forward to match the forward delta (delta threshold is forward delta).

Parameters:: dt – Reference datetime.
Returns:: Delta.