A Research Workflow for Systematic Trading Strategies
Summary
The document sketches a data-driven workflow for developing systematic trading ideas: identify possible market regimes, find predictors from price or fundamental data, combine weaker signals with machine learning or ensembles, and refine rules through position sizing and stop or profit exits. It also suggests finding research inspiration through technical blogs, educational resources, and academic papers.
The replies emphasize that published strategy ideas may lose effectiveness as they become widely known, and that papers can still be useful sources of hypotheses even when their reported approaches do not work reliably in practice. The discussion offers no tested strategy, performance evidence, or detailed safeguards against overfitting, look-ahead bias, and changing market conditions. Its workflow is a starting point for research rather than a validated recipe for building profitable systems.
Key ideas
- Begin by investigating whether market regimes may affect strategy behavior.
- Search price and fundamental data for predictors, then consider combining weak signals.
- Translate signals into trading rules with explicit sizing and exit logic.
- Use papers and practitioner resources to generate hypotheses, not as proof of profitability.
- The suggested process omits detailed validation and bias controls.
Tags
Full text
# Research methodology of systematic strategies # Research methodology of systematic strategies Can someone please share your research methodology of systematic trading strategies? I feel like I am always using the a same data driven procedures over different underlyings and would like to get some inspiration. Here is my simple process 1) determine if there is any market regime pattern 2) Find suitable predictors using price/fundamental data 3) Combine weak predictors with some machine learning/ensembling techniques 4) Refine trading rules with accurate sizing and stop loss/profit. If there is any existing forum or journals that are discussing topoics like this please also share. ## Answer by Richi Wa (score 3) https://quant.stackexchange.com/a/16148 If you like an R related blog with a lot of code that you can use then you shoud look here: http://systematicinvestor.wordpress.com/ Similar in a similar vein but with less code (and I think the authors know each other) is the Blog by David Varadi https://cssanalytics.wordpress.com/ I think these two are an important extension of the ones already mentioned. ## Answer by Quantopik (score 1) https://quant.stackexchange.com/a/16144 It is difficult to find what you need for because if someone shares his knowledge about systematic trading, all the profit of that strategy vanishes in a while theoretically. Anyway, there are a lot of blogs about trading strategies that provide references and guides about that. In my humble opinion, one of the best is QuantStart, that provides a lot of useful codes in python and references about starting in algorithmic trading. Moreover, there is also the Andy Nguyen's website that provides useful guide about what you need for (guides, forum, tutorial, codes in C++,...). There is also the E. P. CHan's blog; take a look at here. He wrote two interesting books about systematic and algo trading: I found those ones really interesting and could be useful to build a basic knowledge to create a trading system. ## Answer by Escachator (score 1) https://quant.stackexchange.com/a/16479 Something I do to find new ideas is read papers in Google Academic related to the topic. Find papers under "quantitative strategy", "trading strategy", etc and you will find interesting things. My experience is that they don't tend to work, but they provide with ideas that you can mix and maybe find something by yourself.
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.