Quant portfolio management is the continuous management of systematic strategies at portfolio level: measuring whether strategies still behave as expected, understanding their interactions, controlling risk and deciding how capital should be distributed between them.
Why portfolio management is different from strategy research
Research asks whether a strategy has enough evidence to be considered investable. Portfolio management asks a different question: given several eligible strategies, how should capital be distributed today, and what should cause that distribution to change?
A strategy can remain individually attractive while becoming a worse fit for the portfolio. Correlations can increase, volatility can change, drawdowns can overlap and several models can become exposed to the same underlying market condition. That is why a quant portfolio process has to look beyond standalone returns.
In a multi-strategy portfolio, the relevant unit of analysis is therefore both the individual model and the system of models sharing the same risk budget.
Continuous strategy monitoring
The first layer is monitoring. Teams need to distinguish ordinary variance from evidence that a strategy is behaving differently from its tested or recent operating range. Useful monitoring can include drawdown state, realised volatility, return distribution changes, execution behaviour, correlation drift and regime context.
Monitoring should not automatically trigger trading. Its job is to identify when the evidence behind a strategy deserves review and to record what changed. A separate strategy drift detection process can make that review more systematic.
Portfolio risk is about interactions
Portfolio risk is not the sum of strategy-level stop losses. The important questions are how strategies interact: whether exposures overlap, whether diversification disappears under stress, whether concentration rises, and whether the same market driver affects several strategies at once.
Capital allocation should connect evidence to constraints
An allocation process should combine strategy evidence with portfolio constraints. That can mean reducing a weight when reliability weakens, limiting an otherwise attractive strategy because concentration is already high, or leaving a weight unchanged when the signal is ambiguous.
The important property is explainability. A professional team should be able to reconstruct what the system saw, which constraints applied, what response was proposed and why.
A repeatable quant portfolio workflow
- Define which strategies are eligible for capital.
- Monitor strategy behaviour and market context continuously.
- Recalculate portfolio-level risk and interactions.
- Generate bounded capital-allocation proposals.
- Apply explicit limits and human review where required.
- Record the decision and measure the subsequent outcome.
Prospect is being built around this operating loop. The objective is not prediction theatre; it is a more consistent and auditable way to connect changing evidence with portfolio decisions.
Initial Prospect deployment: analytics and decision support. Customers retain discretion over capital deployment.