Motor de inferência EVT
- #Rust
A Startup Dimensão Quântica apresenta :
Um motor de inferência de risco de cauda (EVT) aplicado a microestrutura com dados reais.
Pipeline:
• Hill adaptativo
• GPD via PWM (robusto)
• Validação com KS test
• Replay contínuo de eventos
Resultados consistentes:
• ξ ≈ 0.30 – 0.33
• KS ≈ 0.02 – 0.04
• Regime estável
Foco: comportamento sob extremos, não métricas médias.
Quant - EVT - HFT - Risk - Engineering .
“Resultados consistentes ao longo de múltiplos ciclos independentes (evidência abaixo):”
runner@c96fac57f624:~/workspace/Motor_Core$ cargo run --release
Finished `release` profile [optimized] target(s) in 0.03s
Running `target/release/motor_v57`
Cycle 0
Hill Target: 0.3096
PWM Result: 0.3065
KS Statistic: 0.0356
Inference: STABLE REGIME
Cycle 1
Hill Target: 0.3277
PWM Result: 0.3315
KS Statistic: 0.0378
Inference: STABLE REGIME
Cycle 2
Hill Target: 0.3069
PWM Result: 0.2570
KS Statistic: 0.0322
Inference: STABLE REGIME
Cycle 3
Hill Target: 0.2933
PWM Result: 0.2191
KS Statistic: 0.0341
Inference: STABLE REGIME
Cycle 4
Hill Target: 0.3031
PWM Result: 0.2279
KS Statistic: 0.0348
Inference: STABLE REGIME
Cycle 5
Hill Target: 0.2943
PWM Result: 0.2787
KS Statistic: 0.0261
Inference: STABLE REGIME
Cycle 6
Hill Target: 0.3199
PWM Result: 0.3098
KS Statistic: 0.0257
Inference: STABLE REGIME
Cycle 7
Hill Target: 0.3112
PWM Result: 0.1873
KS Statistic: 0.0333
Inference: STABLE REGIME
Cycle 8
Hill Target: 0.3137
PWM Result: 0.3630
KS Statistic: 0.0271
Inference: STABLE REGIME
Cycle 9
Hill Target: 0.2980
PWM Result: 0.4916
KS Statistic: 0.0299
Inference: STABLE REGIME
runner@c96fac57f624:~/workspace/Motor_Core$
