Stochastic Modeling of Scientific Data

Stochastic Modeling of Scientific Data combines stochastic modeling and statistical inference in a variety of standard and less common models such as point processes Markov random fields and hidden Markov models in a clear thoughtful and succinct manner. The distinguishing feature of this work is that in addition to probability theory it contains statistical aspects of model fitting and a variety of data sets that are either analyzed in the text or used as exercises. Markov chain Monte Carlo methods are introduced for evaluating likelihoods in complicated models and the forward backward algorithm for analyzing hidden Markov models is presented. The strength of this text lies in the use of informal language that makes the topic more accessible to non-mathematicians. The combinations of hard science topics with stochastic processes and their statistical inference puts it in a new category of probability textbooks. The numerous examples and exercises are drawn from astronomy geology genetics hydrology neurophysiology and physics. Language: English