Backtesting Frameworks That Accept Custom Market Data
Summary
This discussion considers backtesting platforms for users who need to supply their own market data and use third-party libraries in strategy code. The replies suggest Python-based approaches, including the Python Algorithmic Trading Library, Zipline, and Backtesting.py. They describe options such as importing CSV or OHLC data and using external libraries, including machine-learning and indicator tools. A separate reply mentions a form-based service with built-in strategy signals and instrument coverage.
The exchange is a set of recommendations, not a systematic evaluation. It gives no benchmark results, compatibility checks, licensing comparison, or guidance on data quality, survivorship bias, transaction costs, or execution assumptions. Some platform capabilities may also depend on software versions or integrations, and one recommendation includes a founder’s commercial affiliation. Readers should verify current support for their markets, custom data formats, and dependencies before choosing a tool.
Key ideas
- Custom price data and third-party libraries are important platform criteria for flexible backtesting.
- The replies recommend several Python-based frameworks that can work with imported data.
- The discussion does not compare platforms using benchmarks or controlled examples.
- Users should verify current data, library, and market support and account for backtest assumptions.
Tags
Full text
# Backtesting software with custom data input # Backtesting software with custom data input I was considering to develop a custom backtesting platform for myself. However, I see that it would require some significant time and effort, and the result might not be as initially expected. So I decided to buy a professional backtesting platform. I live outside of the United States and not interested in US stocks, so my top requirement is being able to enter custom price data as input. It could be as a csv file, data source like Google or Yahoo Finance, or some other way. Also, one should be able to use 3rd party libraries in the strategy code, such that machine learning libraries or ta-lib. Please suggest some backtesting software for these requirements. It could be both open-source and commercial. ## Answer by chjortlund (score 3) https://quant.stackexchange.com/a/15833 Have you thought about using "Python for finance"? There is multiple Python librarys to help you getting up to speed, e.g. take a look at Python Algorithmic Trading Library ## Answer by KarolisR (score 3) https://quant.stackexchange.com/a/24493 You can also try Zipline, it's the library used in Quantopian platform. It is opensource and written in python, you can use your own .csv data or built-in yahoo finance data feed. You can of course use any python library you want with it. ## Answer by K3---rnc (score 1) https://quant.stackexchange.com/a/43572 Python backtesting framework Backtesting.py works with any kind of OHLC data and supports arbitrary indicator / machine learning library. ## Answer by ivanftp (score 0) https://quant.stackexchange.com/a/61426 PyInvesting allows you to backtest your investment strategy without writing a single line of code. - Simply fill in a form specifying your backtest details - Create signals using both technical and fundamental data - Backtest your prefered investment strategy (Relative Strength, Fundamentals, Moving Average and Strategic Allocation) - Performance analysis is a breeze with our clean and beautiful user interface - Extensive coverage of instruments (stocks, ETFs, FX and Crypto) across multiple exchanges - Allows you to go live and profit from your investment strategy where you will receive daily email updates about any live orders. Disclaimer: I’m the founder of pyinvesting.com, a backtesting software for stock market investors.
Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.