遍历 Lévy 驱动 SDE 的 Gaussian 准信息准则
Gaussian quasi-information criteria for ergodic Lévy driven SDE
AI总结:
针对高频观测的半参数遍历 Lévy 驱动模型,基于两阶段 Gaussian 准似然函数提出显式的 GQAIC 和 GQBIC 统计量,揭示了联合 GQLF 混合速率结构导致的非标准正则化项,定量阐明了估计精度与采样频率的关系。
摘要:
我们考虑对高频观测的半参数遍历 Lévy 驱动模型的参数系数进行相对模型比较。我们的渐近性基于 Euler-近似类型的完全显式两阶段 Gaussian 准似然函数(GQLF)。为了选择尺度系数和漂移系数,我们通过逐步推断程序提出了显式的 Gaussian 准 AIC(GQAIC)和 Gaussian 准 BIC(GQBIC)统计量。特别地,我们证明了联合 GQLF 的混合速率结构(这在扩散过程中不会出现)导致了尺度系数选择中正则化项的非标准形式,定量地阐明了估计精度与采样频率之间的关系。给出了数值实验以说明我们的理论发现。
英文摘要:
We consider relative model comparison for the parametric coefficients of a semiparametric ergodic Lévy driven model observed at high-frequency. Our asymptotics is based on the fully explicit two-stage Gaussian quasi-likelihood function (GQLF) of the Euler-approximation type. For selections of the scale and drift coefficients, we propose explicit Gaussian quasi-AIC (GQAIC) and Gaussian quasi-BIC (GQBIC) statistics through the stepwise inference procedure. In particular, we show that the mixed-rates structure of the joint GQLF, which does not emerge for the case of diffusions, gives rise to the non-standard forms of the regularization terms in the selection of the scale coefficient, quantitatively clarifying the relation between estimation precision and sampling frequency. Numerical experiments are given to illustrate our theoretical findings.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。