quantdata.dysanos.grid.rndysanos.local_arbitrage#
Exact local-support masks for cost-adjusted option directions.
For direction d equal to +1 (long) or -1 (short), the empirical
candidate function is
arb_d(z, omega) = 1
when every one of the k nearest training observations has
d * dH_omega - c_omega >= -e_omega and at least one has a gain strictly
above e_omega. The tolerance e only absorbs floating-point error;
the economic cost c is subtracted explicitly.
- class quantdata.dysanos.grid.rndysanos.local_arbitrage.LocalArbitrageConfig(*, neighbors=10240, query_batch_size=2048, instrument_batch_size=10, roundoff_multiplier=32.0, state_scale_floor=1e-06)[source]#
Bases:
objectLogical and batching configuration for the exact local mask.
- class quantdata.dysanos.grid.rndysanos.local_arbitrage.LocalArbitrageProgress(*, split, processed_states, total_states, long_only_candidate_states, short_only_candidate_states, both_direction_candidate_states)[source]#
Bases:
objectCumulative disjoint candidate-state counts during mask valuation.
- quantdata.dysanos.grid.rndysanos.local_arbitrage.cost_adjusted_direction_candidates(neighbor_returns, query_cost, neighbor_roundoff)[source]#
Return long/short candidate flags with shape
(B, 2, J).
- quantdata.dysanos.grid.rndysanos.local_arbitrage.pack_direction_mask(mask)[source]#
Pack the option axis of a bool
(..., 2, J)candidate mask.
- quantdata.dysanos.grid.rndysanos.local_arbitrage.unpack_direction_mask(packed, *, instruments)[source]#
Unpack a uint8
(..., 2, ceil(J/8))mask on its Torch device.
- quantdata.dysanos.grid.rndysanos.local_arbitrage.valuate_local_arbitrage_mask(*, reference_state, reference_returns, reference_roundoff, query_state, query_returns, query_cost, query_roundoff, config, split, leave_one_out, candidate_sink=None, progress=None, progress_interval_seconds=30.0)[source]#
Evaluate the training-fitted exact kNN
arb(z; omega)function.