# Model card — sv (stochastic volatility) — UNSUPPORTED, not scored

# 1. Identity

• model_id: sv · family: score-driven (non-GARCH) in the registry grouping · tier: A · status: unsupported · impl: none · distributions: [] (registry note: "simulation-based estimation not feasible at the rolling re-estimation cadence of this benchmark; not scored")
• Seminal citation: kim1998stochastic (Kim, Shephard & Chib 1998). Related: taylor1994modeling, taylor2008modelling (origin), jacquier1994bayesian (Bayesian SV), harvey1994multivariate (QML via the Kalman filter).

# 2. Definition (literature)

Log-SV: $r_t = \mu + \exp(h_t/2),z_t$, $h_t = \alpha + \phi,h_{t-1} + \sigma_\eta,\eta_t$, $(z_t, \eta_t)$ i.i.d. Gaussian (possibly correlated for leverage); the variance is latent, so the likelihood is an integral over the volatility path.

# 3–6. Parameter space, estimation, forecasting

Not implemented. Estimation requires MCMC/particle filtering or QML on $\log r_t^2$ (harvey1994multivariate), and forecasts require filtering the latent state.

# 7. Implementation / why not scored

Rolling re-estimation every 5 days over ≈ 42 instruments × 3,700 origins makes simulation-based estimation infeasible within the compute budget (registry note, D-11); no version-locked library in the stack. Listed in the matrix as "extension (optional, 'if feasible')".

# 8–9. Limitations / footprint

Not applicable; compute class "simulation/Bayesian" (brief).

# 10. Literature expectations

kim1998stochastic compare SV with GARCH and find SV competitive in likelihood terms; no cross-asset multi-horizon forecasting benchmark in the matrix.