Costs & Commissions - ZIPLIME

Every fill pays a commission and a slippage model, and commission reaches the cost basis and realised P&L, not only cash. What a run charges by default depends on how you start it.

Defaults

	`run_simulation(...)`	`ziplime run`
Equities	`PerShare(cost=0.001)` — $0.001 per share, no minimum	`PerShare(cost=0.0)` — no commission unless `--commission-per-share`
Futures	`PerContract(cost=0.85)` plus exchange fees by root symbol	same
Options	`PerOptionContract()` — $0.65 per contract plus$ 0.05 exchange fee	same
Bonds	`PerBondTurnover(cost=0.0003)` — 3 bps of turnover	same
Equity slippage	`FixedBasisPointsSlippage(basis_points=5.0, volume_limit=0.1)`	`--slippage-bps 5.0`, same volume cap
Futures slippage	`VolatilityVolumeShare(volume_limit=0.05)`	same

ziplime run charges no equity commission by default. Pass --commission-per-share for any run whose numbers you intend to trust.

Commission models

from ziplime.finance.commission import PerShare, PerDollar, PerContract, PerOptionContract
from ziplime.finance.commission.per_trade import PerTrade

Model	Asset class	Charges
`PerShare(cost, min_trade_cost)`	Equities	Per share, with an optional minimum per trade
`PerDollar(cost)`	Equities	A fraction of traded value
`PerTrade(cost)`	Any	A flat fee per trade
`PerContract(cost, exchange_fee, min_trade_cost)`	Futures	Per contract, plus exchange fees
`PerFutureTrade(cost)`	Futures	A flat fee per futures trade
`PerOptionContract(cost, exchange_fee, min_trade_cost)`	Options	Per contract, plus exchange fees
`PerBondTurnover(cost)`	Bonds	A fraction of turnover
`NoCommission()`	Any	Nothing — a control run only

Slippage models

Model	Behaviour
`FixedBasisPointsSlippage(basis_points=5.0, volume_limit=0.1)`	Fixed price impact; each fill is capped at `volume_limit` of the bar’s volume. `basis_points=0` is a control run that is still volume-capped
`VolatilityVolumeShare(volume_limit=...)`	Futures impact scaled by volatility and volume share
`NoSlippage()`	Fill at the bar price with no impact

Setting costs on a run

from ziplime.finance.commission import PerDollar
from ziplime.finance.slippage.fixed_basis_points_slippage import FixedBasisPointsSlippage

result = await run_simulation(
    ...,
    equity_commission=PerDollar(cost=0.001),
    equity_slippage=FixedBasisPointsSlippage(basis_points=5.0, volume_limit=0.1),
)

ziplime run -f sma.py -b demo -s AAPL,MSFT --start-date 2020-01-01 --end-date 2025-12-31 \
  --commission-per-share 0.005 --slippage-bps 5

Costs belong to the run, not to the strategy. context.set_commission(...) and context.set_slippage(...) exist for Zipline compatibility; prefer the run’s parameters so the cost model is visible next to the result.

Vectorized runs

The vectorized kernel applies the same commission and slippage models. A cost model the kernel can’t apply stops the run by name — unpaid costs would otherwise show up as a higher return, not as an error.

A realistic cost model narrows the gap between backtest and live results, but doesn’t close it. No backtest reproduces the future.