Estimating Strategy Capacity Through Risk, Position Size, and Market Impact
Summary
The discussion addresses how to estimate a trading strategy’s capacity using historical data and market depth. The question notes that standard broker market-impact tools may not capture strategies with complex entries and exits or those whose opportunities depend on specific liquidity events. It offers no numerical capacity analysis or empirical comparison of models.
The response recommends working from investment objectives toward impact estimates. First define required returns, risk tolerance, and risk controls; these inform approximate position sizes. Then select suitable asset classes, markets, and holding periods, develop the strategy, and estimate the market impact of its orders given those choices. The central point is that impact depends on order size and context, so gross returns cannot be meaningfully compared with impact in isolation. The advice is a high-level planning sequence, not a formula for calculating capacity. It also says that assessing a particular strategy’s expected net performance requires more information about the strategy and its intended trading scale.
Key ideas
- Market impact depends heavily on order size and the strategy’s trading context.
- Define return objectives, risk tolerance, and risk controls before estimating position size.
- Choose markets and holding periods that fit the required risk and return profile.
- Estimate order impact after specifying the strategy and approximate positions.
- The discussion provides a planning framework rather than a quantitative capacity model.
Tags
Full text
# How do you estimate the capacity of a strategy from historical data? # How do you estimate the capacity of a strategy from historical data? What are some good ways to estimate the capacity of a strategy from historical data (including full market depth)? Obviously, a naive approach is that you want the strategy's returns to exceed its market impact. PBs mostly provide their institutional clients with software packages that estimate market impact for large orders. However, many strategies have unique and more intricate entries/exit than large-order-buy-and-hold. Moreover, there are strategies that are low-capacity and contingent on market depth events, both of which I'm guessing require very different underlying assumptions than the market impact models used for large orders. ## Answer by Matt Wolf (score 3) https://quant.stackexchange.com/a/7110 I believe you approach this whole issue from the wrong end. - Market impact is a huge function of the size of your orders and therefore you cannot start to ask what strategies outperform certain market impact. - Instead you should start to think about required returns and associated risk tolerance. Also define prudent risk management rules. From that results an idea about the approximate trading size you are willing to put at risk. - After you clearly defined your risk/reward you will start looking which asset classes, market, holding periods, most likely satisfy your requirements. - After that you fine-tune your strategy approach and start thinking about particular strategies. - And when you have done all that you start thinking about market impact on your orders as a function of all the above plus your estimated position sizes. Without any other information on your end I find it impossible to help you further. I am not asking for anything related to your strategy but its very hard to make an educated guess whether your particular strategy's expected gross return exceeds transaction related cost, including market impact. You rejected to volunteer any other information with a condescending attitude. Maybe others appreciate your particular approach to asking for help more than I do.
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.