Handbook of Biosurveillance ISBN 0-12-369378-0
Modeling the likelihood function, P(E\H), involves specifying parameters within the model. For example, in the simplest case that E and H each consist of one binary variable, then two parameters are needed to specify the likelihood function. P(H) represents the prior probability of the hypothesis itself (e.g., the probability of influenza).The parameters and hypothesis priors may be estimated in a variety of ways, including the use of previously available training data, the literature, expert opinion, and their combination. Section 4.1 illustrates how priors are estimated from a combination of information sources.
Was this article helpful?