Reduced-complexity smoothing for hidden Markov models

We investigate approximate smoothing schemes for a class of hidden Markov models (HMM), namely, HMMs with underlying Markov chains that are nearly completely decomposable. The objective is to obtain substantial computational savings. Our algorithm can not only be used to obtain aggregate smoothed estimates, but can be used to obtain systematically approximate full-order smoothed estimates with computational savings, unlike many of the aggregation methods proposed earlier.