Combined parameter and state estimation in simulation-based ltering

@inproceedings{LiuCombinedPA,
  title={Combined parameter and state estimation in simulation-based ltering},
  author={Jane Liu and Mike West}
}
We discuss simulation-based sequential analysis { or particle ltering { in dynamic models, with a focus on sequential Bayesian learning about time-varying state vectors and xed model parameters simultaneously. This includes a general approach that combines old ideas of smoothing using kernel methods with newer ideas of auxiliary particle ltering. We highlight speci c smoothing approaches that have interpretation as adding \arti cial evolution noise" to xed model parameters at each time point to… CONTINUE READING