Hidden Markov Models for Time Series: An Introduction Using R (2nd Edition)
Abdolvahab Khademi · Journal of Statistical Software · 2017
The distribution of estimable parameters of observed data in some stochastic processes may be influenced by another yet unobservable process called the parameter process.Such a parameter process can be serially dependent.A tractable mathematical model to account for serially dependent processes is the Markov chain process, which can be applied to both observed and unobserved processes.A Markov chain process used to model an unobserved serially dependent process is commonly called hidden Markov model (HMM), which is used in fields as diverse as signal processing, machine learning (e.g., handwriting recognition), environmental sciences, economics, and longitudinal data in social or medical sciences.Given the importance and widespread use of HMM, several books and references have been published for students, practitioners, and researchers.Hidden Markov Models for Time Series: An Introduction Using R (2nd Edition) is a book on HMM dedicated to time series data for researchers who cannot use a standard time series model for their data, including researchers in animal behavior, epidemiology, finance, hydrology, and sociology, among others.