Linearized Filtering of Affine Processes Using Stochastic Riccati Equations

We consider an affine process $ X $ which is only observed up to an  additive white noise, and we ask for the law of $ X_t $, for some  $ t > 0 $, conditional on observations up to time $ t $. This is  a possibly high... [ view full abstract ]

We consider an affine process $ X $ which is only observed up to an  additive white noise, and we ask for the law of $ X_t $, for some  $ t > 0 $, conditional on observations up to time $ t $. This is  a possibly high dimensional filtering problem which is not even locally approximately Gaussian, whence essentially only  particle filtering methods remain. In this  work we present an efficient filter by solving a system of stochastic generalized Riccati differential equations. The efficiency is illustrated with filtering affine stochastic variance processes from discrete price observations.