Reading assignment:
The biggest problem with hidden Markov models is that there are quite a few different sources of information and every one uses its own notation. Sometimes the notations overlap in very confusing ways.
To get around this, I put together a fairly complete document on HMMs, which you may download here. It is pretty notation-heavy, but if you have problems with the notation (which is actually not bad once you take the time to figure out what's going on), then you can talk to me or one of your classmates.
Durbin et al Chapter 3.1 - 3.5. An excellent introduction to Markov & Hidden Markov models.
Ewens and Grant Chapter 12. HMMs from another perspective, possibly easier to read.
Sean Eddy's clear and succinct primer in Nature Biotechnology, 2004.
A nice website on HMMs, with models (watch for changing notation though): University of Leeds
The biggest problem with hidden Markov models is that there are quite a few different sources of information and every one uses its own notation. Sometimes the notations overlap in very confusing ways.
To get around this, I put together a fairly complete document on HMMs, which you may download here. It is pretty notation-heavy, but if you have problems with the notation (which is actually not bad once you take the time to figure out what's going on), then you can talk to me or one of your classmates.
Durbin et al Chapter 3.1 - 3.5. An excellent introduction to Markov & Hidden Markov models.
Ewens and Grant Chapter 12. HMMs from another perspective, possibly easier to read.
Sean Eddy's clear and succinct primer in Nature Biotechnology, 2004.
A nice website on HMMs, with models (watch for changing notation though): University of Leeds