Score-Based Parameter Estimation for a Class of Continuous-Time State Space Models

08/18/2020 āˆ™ by Alexandros Beskos, et al. āˆ™ 0 āˆ™

We consider the problem of parameter estimation for a class of continuous-time state space models. In particular, we explore the case of a partially observed diffusion, with data also arriving according to a diffusion process. Based upon a standard identity of the score function, we consider two particle filter based methodologies to estimate the score function. Both methods rely on an online estimation algorithm for the score function of š’Ŗ(N^2) cost, with Nāˆˆā„• the number of particles. The first approach employs a simple Euler discretization and standard particle smoothers and is of cost š’Ŗ(N^2 + NĪ”_l^-1) per unit time, where Ī”_l=2^-l, lāˆˆā„•_0, is the time-discretization step. The second approach is new and based upon a novel diffusion bridge construction. It yields a new backward type Feynman-Kac formula in continuous-time for the score function and is presented along with a particle method for its approximation. Considering a time-discretization, the cost is š’Ŗ(N^2Ī”_l^-1) per unit time. To improve computational costs, we then consider multilevel methodologies for the score function. We illustrate our parameter estimation method via stochastic gradient approaches in several numerical examples.



There are no comments yet.


page 1

page 2

page 3

page 4

This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.