Simulator Controls
Engine Insights — auto-generated for this node
Solvency Surface — every node in the database
Each cell is 1,000 fully simulated retirement paths. Click-drag to zoom.
Modeling Framework & Historical Calibration
Presets are calibrated on market history spanning 2005 → present, capturing the 2008 GFC, the QE decade, and the 2021–23 inflationary regime. Parameters are estimated via Maximum Likelihood (jump-diffusion) and OLS (mean-reversion) solvers, with convergence checks before a preset is admitted.
Merton Jump-Diffusion
dSₜ/Sₜ = (μ−λk)dt + σdWₜ + (J−1)dNₜ
GBM growth plus Poisson-driven discontinuous crashes, fitted to MSCI World (URTH) total returns.
Cox–Ingersoll–Ross
drₜ = κ(θ−rₜ)dt + σ√rₜ dWₜ
Square-root diffusion keeps nominal yields non-negative; parameterized to the Effective Fed Funds Rate (FRED).
Ornstein–Uhlenbeck
dπₜ = κ(μπ−πₜ)dt + σdWₜ
Mean-reverting CPI dynamics with central-bank feedback, fitted to rolling 12M YoY CPI (CPIAUCSL proxy).
Expert Parameters (World Baseline)
- Frictional drag: 25 bps annual portfolio fee, applied monthly.
- Capital gains tax: 15% flat, computed dynamically on rebalancing & distributions.
- Rate floor: short rate restricted to ≥ 0.1% to avoid invalid roots.
- Reproducibility: every node seeded (seed = 42); 1,000 correlated paths each.
Cloud Calibration Pipeline
- Upload raw historical CSVs via
/api/v1/calibrate(multipart). - Engine runs MLE / OLS parameter estimation on cloud hardware.
- Mathematical convergence checks validate the fit.
- A bespoke regional preset JSON is returned to your tenant.
SaaS API Console
Payloads below regenerate live as you move the sliders.
| Capability | This Demo | Cloud API |
|---|---|---|
| Initial balance | $1.0M fixed | Any amount |
| Horizon & scenarios | 50y · 1,000 | Fully configurable |
| Regional presets | World only | USA · Brazil · Eurozone · Japan |
| Custom calibration (CSV upload) | — | ✓ |
| Tax / fee / guardrail parameters | Fixed | Per-request |
| Usage tracking & persistent state | — | ✓ |
▸ Example response payload (illustrative)
{
"status": "success",
"job_id": "sim_9f3K2x",
"compute_ms": 812,
"scenarios": 1000,
"probability_of_success": [1.0, 0.998, "…", 0.731],
"median_final_balance": 412880.55,
"percentiles": { "p5": 0, "p50": 412880.55, "p95": 3811204.19 },
"download_url": "https://api.aethel.com/v1/results/sim_9f3K2x.parquet"
}