Spaces:
Running
Running
File size: 6,680 Bytes
5d18614 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 | """Next-open execution, explicit costs, and accounting reconciliation."""
from __future__ import annotations
from datetime import UTC, date, datetime, timedelta
from decimal import Decimal
from pathlib import Path
from tradebench.data.query import PitQueryService
from tradebench.data.sample_dataset import ASSET_LIQUID, SAMPLE_DATASET_VERSION
from tradebench.episodes.models import EpisodeManifest
from tradebench.execution.costs import (
CostModel,
compute_fee_total,
compute_slippage_bps,
fill_price_from_open,
)
from tradebench.execution.engine import (
OrderIntent,
advance_trading_session,
portfolio_market_value,
)
from tradebench.ledger.events import (
DayAdvanced,
DividendApplied,
LedgerEvent,
OrderFilled,
)
from tradebench.ledger.projector import project
def _manifest(sample_dataset_root: Path) -> EpisodeManifest:
path = (
sample_dataset_root
/ "catalog"
/ SAMPLE_DATASET_VERSION
/ "episode_manifests"
/ "sample_ep_001.json"
)
return EpisodeManifest.model_validate_json(path.read_text(encoding="utf-8"))
def _last_close(pit: PitQueryService, asset_id: str, session_date: date) -> Decimal:
df = pit.get_bars([asset_id], session_date, lookback_days=1)
row = df.iloc[-1]
return Decimal(str(row["close"]))
def test_fills_use_next_session_open_not_decision_close(
pit_query_service: PitQueryService,
sample_dataset_root: Path,
) -> None:
"""Buy decided on ``t`` must fill at ``t+1`` open (adjusted), not at ``t`` close."""
pit = pit_query_service
manifest = _manifest(sample_dataset_root)
dec = date(2020, 1, 2)
exe = date(2020, 1, 3)
model = CostModel(
commission_floor=Decimal("1"),
fee_bps=Decimal("10"),
slippage_bps_cap=Decimal("50"),
)
prior: list[LedgerEvent] = [
DividendApplied(
event_id="boot",
episode_id="ep",
event_time=datetime(2020, 1, 1, 12, 0, 0, tzinfo=UTC),
asset_id=ASSET_LIQUID,
cash_credited=Decimal("100000"),
),
]
step = advance_trading_session(
pit=pit,
manifest=manifest,
episode_id="ep",
decision_date=dec,
execution_date=exe,
prior_events=prior,
new_orders=[
OrderIntent(
client_order_id="b1",
asset_id=ASSET_LIQUID,
side="buy",
quantity=10,
),
],
cost_model=model,
base_event_time=datetime(2020, 1, 2, 16, 0, 0, tzinfo=UTC),
)
fills = [e for e in step if isinstance(e, OrderFilled)]
assert len(fills) == 1
fill = fills[0]
dec_close = _last_close(pit, ASSET_LIQUID, dec)
exe_open = Decimal(str(pit.get_bars([ASSET_LIQUID], exe, 1).iloc[-1]["open"]))
assert dec_close == Decimal("100")
assert exe_open == Decimal("99.75")
bar_dec = pit.get_bars([ASSET_LIQUID], dec, 1).iloc[-1]
dv = Decimal(str(bar_dec["dollar_volume"]))
slip = compute_slippage_bps(
model,
order_notional=Decimal(fill.quantity) * exe_open,
trailing_dollar_volume=dv,
)
expected_px = fill_price_from_open(exe_open, "buy", slip)
assert fill.avg_fill_price == expected_px
assert fill.avg_fill_price != fill_price_from_open(dec_close, "buy", slip)
def test_cash_debits_include_commission_and_bps_fees(
pit_query_service: PitQueryService,
sample_dataset_root: Path,
) -> None:
pit = pit_query_service
manifest = _manifest(sample_dataset_root)
model = CostModel(
commission_floor=Decimal("2"),
fee_bps=Decimal("25"),
slippage_bps_cap=Decimal("100"),
)
prior: list[LedgerEvent] = [
DividendApplied(
event_id="boot",
episode_id="ep",
event_time=datetime(2020, 1, 1, 12, 0, 0, tzinfo=UTC),
asset_id=ASSET_LIQUID,
cash_credited=Decimal("100000"),
),
]
step = advance_trading_session(
pit=pit,
manifest=manifest,
episode_id="ep",
decision_date=date(2020, 1, 2),
execution_date=date(2020, 1, 3),
prior_events=prior,
new_orders=[
OrderIntent(
client_order_id="b1",
asset_id=ASSET_LIQUID,
side="buy",
quantity=5,
),
],
cost_model=model,
base_event_time=datetime(2020, 1, 2, 16, 0, 0, tzinfo=UTC),
)
fill = next(e for e in step if isinstance(e, OrderFilled))
fill_notional = fill.avg_fill_price * Decimal(fill.quantity)
expected_fees = compute_fee_total(model, fill_notional=fill_notional)
assert fill.fees == expected_fees
st = project([*prior, *step])
assert st.cash >= Decimal("0")
def test_scripted_episode_reconciles_after_each_day_advanced(
pit_query_service: PitQueryService,
sample_dataset_root: Path,
) -> None:
"""``DayAdvanced.portfolio_market_value`` matches cash + marks from projection."""
pit = pit_query_service
manifest = _manifest(sample_dataset_root)
model = CostModel(
commission_floor=Decimal("1"),
fee_bps=Decimal("10"),
slippage_bps_cap=Decimal("30"),
)
events: list[LedgerEvent] = [
DividendApplied(
event_id="boot",
episode_id="ep",
event_time=datetime(2020, 1, 1, 12, 0, 0, tzinfo=UTC),
asset_id=ASSET_LIQUID,
cash_credited=Decimal("100000"),
),
]
days = [
(date(2020, 1, 2), date(2020, 1, 3)),
(date(2020, 1, 3), date(2020, 1, 6)),
]
for dec, exe in days:
step = advance_trading_session(
pit=pit,
manifest=manifest,
episode_id="ep",
decision_date=dec,
execution_date=exe,
prior_events=events,
new_orders=[
OrderIntent(
client_order_id=f"buy-{exe}",
asset_id=ASSET_LIQUID,
side="buy",
quantity=1,
),
],
cost_model=model,
base_event_time=datetime.combine(dec, datetime.min.time()).replace(
tzinfo=UTC,
)
+ timedelta(hours=16),
)
events.extend(step)
adv = next(e for e in step if isinstance(e, DayAdvanced))
st = project(events)
closes = {
ASSET_LIQUID: _last_close(pit, ASSET_LIQUID, exe),
}
recomputed = portfolio_market_value(st, closes)
assert adv.portfolio_market_value == recomputed
|