portfolio-optimizer

Portfolio optimization under risk constraints requires identifying optimal strategy allocations across multiple windows like 7d, 30d, and 90d, then generating a frontier of risk/return tradeoffs with final recommended weights that respect concentration limits and turnover caps.

Regime-aware rebalancing adapts portfolio allocation weights based on risk-on, risk-off, or neutral regime signals, applying tuned parameters across multiple windows to generate recommended allocations that align with current market conditions.

Tuning strategy parameters across multiple time windows involves applying the optimizer across 7d, 30d, and 90d windows to discover optimal settings, then producing a frontier of risk/return tradeoffs with confidence scores and recommended weights.

Max/min weight limits, concentration limits, and turnover caps can be enforced during portfolio rebalancing through predefined risk constraints, while deterministic fallback policies and parameter bounds ensure safety rules are respected.

Generating an efficient frontier for a universe of strategies requires running the optimizer across multiple windows and regime signals to produce risk/return tradeoffs, recommended weights, tuned parameters, and confidence metrics in a structured JSON-like payload.

The portfolio optimizer enforces safety rules and deterministic fallback policies via published contracts like parameter-bounds and fallback-policy, ensuring constraints are respected and producing outputs with frontier, confidence, and generated_at fields.