egutierrez
47fac22230
- .claude/CLAUDE.md - .claude/commands/subagentes.md - .claude/rules/INDEX.md - .mcp.json - bash/functions/cybersecurity/analyze_dns.md - bash/functions/cybersecurity/audit_http_headers.md - bash/functions/cybersecurity/audit_ssh_config.md - bash/functions/cybersecurity/check_firewall.md - bash/functions/cybersecurity/detect_suspicious_users.md - bash/functions/cybersecurity/encrypt_file.md - ... Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
43 lines
1.5 KiB
Markdown
43 lines
1.5 KiB
Markdown
---
|
|
name: estimate_hawkes
|
|
kind: function
|
|
lang: py
|
|
domain: datascience
|
|
version: "1.0.0"
|
|
purity: pure
|
|
signature: "def estimate_hawkes(arrivals: list[int], max_lag: int = 30) -> dict"
|
|
description: "Estima parámetros de un proceso Hawkes (alpha, beta, branching_ratio) desde la autocorrelación de arrivals ajustando una exponencial decreciente sobre la ACF."
|
|
tags: [estimation, hawkes, stochastic-process, microstructure, timeseries, pendiente-usar]
|
|
uses_functions: []
|
|
uses_types: []
|
|
returns: []
|
|
returns_optional: false
|
|
error_type: ""
|
|
imports: [numpy, scipy]
|
|
params:
|
|
- name: arrivals
|
|
desc: "lista de conteos de eventos por periodo (ej: [0, 1, 3, 2, 0, 1, ...] eventos por tick). Reflect actividad temporal."
|
|
- name: max_lag
|
|
desc: "numero maximo de lags para calcular autocorrelacion (tipico: 30). Mayor = mas precision pero mas ruido."
|
|
output: "dict con {alpha, beta, branching_ratio, acf} estimados parametros del proceso Hawkes"
|
|
tested: false
|
|
tests: []
|
|
test_file_path: ""
|
|
file_path: "python/functions/datascience/datascience.py"
|
|
---
|
|
|
|
## Ejemplo
|
|
|
|
```python
|
|
arrivals = [0, 1, 3, 2, 0, 1, 4, 2, 1, 0] * 10
|
|
result = estimate_hawkes(arrivals, max_lag=10)
|
|
# {'alpha': 0.312, 'beta': 0.874, 'branching_ratio': 0.357, 'acf': [...]}
|
|
```
|
|
|
|
## Notas
|
|
|
|
Ajusta la función `a * exp(-b * lag)` sobre los lags 1..max_lag de la ACF usando `curve_fit` de scipy.
|
|
Si el primer lag de la ACF es <= 0.01 (sin autocorrelación), retorna alpha=0, beta=1, branching_ratio=0.
|
|
El branching_ratio = alpha/beta; si se acerca a 1, el proceso es explosivo.
|
|
Función pura: requiere numpy y scipy instalados.
|