Skip to main content
Trading executed by a predefined algorithm (strategy) rather than manually.
A set of rules that drives trading decisions — when to buy and when to sell.
Testing a strategy against historical data to see how it would have performed in the past.
Common strategy types: managing a basket of instruments, trading a pair of assets against each other, or following market momentum. Ziplime supports any strategy type.
A reference used to compare a strategy’s return — for example, a market index.
A backtesting error where a strategy uses data that didn’t exist yet at the moment of the trade, making the result unreliable.
Tuning parameters so closely to past data that the result looks great historically but fails to generalize going forward.
The difference between the expected and the actual execution price of a trade.
A decline in portfolio value from its peak; maximum drawdown is the deepest such decline.
Standard strategy-quality metrics that express risk-adjusted return and performance relative to the market.
Stored history of changes to a strategy’s code — what changed and when.
An open-source backtesting engine, formerly the core of the Quantopian platform, and the foundation of Ziplime’s engine.
A well-known (now defunct) algorithmic trading platform built on Zipline.
Ziplime’s internal service that supplies market data to the platform.
A shared authentication system — users stay signed in across Limex products.
Lime Trading’s programmatic REST API for order execution; also a standalone product for users who build their own trading systems. See lime.co/rest-api-python-sdk.
A separate Lime Trading product offering direct market access via FIX for technical teams that need minimal latency. See lime.co/lime-direct-api-and-fix-protocol.
Trading where the outcome depends on fractions of a second.
Posting both buy and sell quotes to earn the spread between them; requires speed and infrastructure.
Company financial metrics: statements, valuation multiples.
Non-traditional data sources that support trading decisions.
A model where basic access is free and advanced features are paid.
A cloud model where resources scale automatically with load, with no servers to manage individually.
The strategy languages used by MetaTrader, Quik, and TradingView respectively; Ziplime’s AI converts them to Python.
An algorithmic trading platform and its open-source engine; strategies can be migrated to Ziplime.