Time Varying Maximum Likelihood

Abstract
We consider the problem of tracking an unknown time varying parameter that characterizes the probabilistic evolution of a sequence of independent observations.
To this aim, we propose a novel stochastic gradient descent-based recursive scheme in which the gradient is given by the score of the log-likelihood with respect to the parameter of interest.
We then prove that the scheme converges to a neighborhood of the true time varying parameter.

Organizzazione
Alberto Lanconelli