Features for quant research and backtesting

Build and test trading strategies in Python, from loading market data and calculating signals to simulating orders, analyzing trades, and checking results out of sample. VectorBT PRO lets you compare assets, parameter settings, and time periods in the same research workflow, using your own pandas data and trading rules.

See the features in action through working Python examples, interactive plots, and real backtest results. Start with a strategy idea and explore the tools to develop it further.

Find your starting point

What do you want to do?	Where to start
Turn buy and sell signals, order sizes, or custom rules into a backtest	Backtesting covers simulation, execution, stops, costs, and portfolio accounting.
Prepare data and build a strategy's inputs	Data covers sources, storage, and timeframes. Indicators covers calculations, signals, and machine learning labels.
Understand trades, returns, drawdowns, and what drives a result	Analysis covers metrics, trade behavior, charts, and reports.
Compare settings or allocations and test them on unseen data	Optimization covers parameter searches, walk-forward validation, robustness, and portfolio weights.
See an entire strategy put together	Strategies covers pairs trading, cross-sectional ranking, and portfolios of strategies.

Research across assets, parameters, and time

Start with one strategy, then compare its settings across symbols and time periods. Labeled results keep each combination attached to its signals, trades, and metrics, so you can select one backtest for inspection or compare the whole experiment.

You can calculate signals in advance or make decisions as the simulation runs, using current cash, positions, and previous fills. The Backtesting engine shows the available inputs, and Event-driven backtesting shows how to add your own stateful logic.

As the research grows, Performance covers compiled calculations, optional Rust support, parallel execution, and benchmarks. The Toolkit helps with array operations, reusable configuration, saving results, and automation. AI adds documentation search, chat, and tools for your coding assistant.

Explore all feature areas

VBT builds on the features of open-source vectorbt and extends them in the areas below. Already use vectorbt? The VectorBT PRO vs vectorbt comparison explains what PRO adds and how to move your code.

Data

Connect, store, update, transform, and combine local and remote market data

Indicators and signals

Compute indicators, build your own, stream updates, and turn them into signals

Backtesting

Model orders, signals, leverage, stops, limits, cash flows, and simulation callbacks

Analysis

Measure returns and risk, dissect trades, chart results, and study market events

Optimization and validation

Search parameters, validate out of sample, test robustness, and allocate across assets

Strategies

Follow worked strategies that combine data, signals, simulation, and validation end to end

Performance

Scale computation with chunking, caching, JIT compilation, and parallel execution

Research toolkit

Use labeled arrays, pandas extensions, persistence, automation, and developer tools

AI

Get help with Python backtests, search VBT knowledge, and connect your coding assistant

To keep pages concise, only a selection of the most interesting features from each release is highlighted. Full release notes are available exclusively to subscribers. If you are on the private website, navigate to Getting started → Release notes.

Tags indicate releases where features were introduced for the first time. Please note that most features are continuously updated, so the following examples are intended to be run with the latest version of VBT installed ✍️

Import required by code examples

from vectorbtpro import *