Comparing Capital Use with Time-in-Market and Financing Costs
Summary
The document asks how to compare strategies whose capital usage differs over time. It contrasts strategies with stable allocations against an event-triggered strategy whose invested capital varies, noting that return as a percentage of capital is easier to compare when allocations remain roughly constant. Equal Sharpe ratios do not by themselves resolve the question of capital efficiency when exposure levels and durations differ.
The answer identifies financing costs as one consequence of keeping capital deployed for more of the time, and suggests examining return per unit of time in the market. It also mentions reporting the fraction of time a strategy is invested, while cautioning that it does not know of a standard metric and has not seen the proposed measure widely used. The discussion is conceptual: it provides no formula, data, or adjustment for costs, leverage, capacity, or risk differences. Time-in-market can therefore inform a comparison but is not presented as a complete ranking method.
Key ideas
- Return as a percentage of capital is most directly comparable when capital allocations are fairly stable.
- Equal Sharpe ratios do not capture differences in how much capital is deployed or for how long.
- Financing costs may rise when a strategy keeps capital invested for a greater share of the time.
- Return per unit of time in the market and the fraction of time invested are suggested descriptive measures, not established universal rankings.
Tags
Full text
# Capital efficiency of event triggered strategy # Capital efficiency of event triggered strategy Let's assume we have two long short equity strategies `A`, `B` with a Sharpe ratio of `2` each. Ignoring scalability, trading costs, turnover etc. We want to simply compare how efficient two strategies are using their capital. Strategy `A` returns 10% on it's capital yearly and strategy `B` returns 1% on it's capital yearly. This could be possible if strategy `B` just has very low volatility. Obviously `A` is better because its using it's capital more efficiently. This calculation is very simple when we are comparing % returns. But % returns only makes sense of the capital allocation is roughly constant i.e. we have for example a constant `10m $` invested in each strategy. A third strategy `C` - also with a Sharpe ratio of `2` - has highly non-constant capital allocation let's say between `1m $` and `10m $` invested. It wouldn't make sense to keep strategy `C` at a constant allocation because of the way it trades. Let's say for example it reacts to events that unevenly distributed. How can I compare how efficiently the strategies are using their capital and rank them? Are there any established metrics for that? ## Answer by user42108 (score 0, accepted) https://quant.stackexchange.com/a/60352 "How can I compare how efficiently the strategies are using their capital and rank them? Are there any established metrics for that?" The strategies will incur financing costs when using capital so those which are invested for a higher proportion of the time will incur higher financing costs. You could also look at return per unit of time in the market. But I'm not aware of any standard metrics and I haven't seen this used in practice, though I have seen quant strategies where the backtest reports % of time in the market.
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.