quantopian_templates

Templates for Quantopian

Best Practices for a Fast Algorithm

Access context as infrequently as possible.
Bunch all of the algorithms data look ups:
- data.history
- data.current
- data.can_trade
- data.is_stale
Only record data at the end of each day.
Access account and portfolio data only when needed (calculated daily or on demand).
Logging is limited to 2 logs per call of initialize and handle_data.

Factors

1. Slippage Models

Slippage must be defined in the initalize method. If you do not specify a slippage model it defaults to:

# You can take up to 2.5% of the minute's trading volume
VolumeShareSlippage(volume_limit=0.25, price_impact=0.1)

To set slippage, use:

# set_slippage can take FixedSlippage, VolumeShareSlippage, or a custom slippage model.
set_slippage() <-- slippage.FixedSlippage or slippage.VolumeShareSlippage

slippage.VolumeShareSlippage(volume_limit, price_impact)

volume_limit: limits the proportion of volume that your order can take up per bar.
price_impact: constant (defaults to 0.1) defines how large of an impact your order will have on the backtesters price calculation.

slippage.FixedSlippage(spread)

spread: Typical bid/ask spread

2. Commission Models

You can set commission by trade or share

set_commission() <-- commission.PerShare or commission.PerTrade