Weight Of Evidence

influence of evidence on the decision of a judicial instance

In statistics, the use of Bayes factors is a Bayesian alternative to classical hypothesis testing. Given a model selection problem in which we have to choose between two models M1 and M2, on the basis of a data vector x. The Bayes factor K is given by where is called the marginal likelihood for model i. This is similar to a likelihood-ratio test, but instead of maximising the likelihood, Bayesians average it over the parameters. Generally, the models M1 and M2 will be parametrised by vectors of parameters θ1 and θ2; thus K is given by
See more at Wikipedia.org...

This phrase is used to signify that the proof on one side, of a cause is greater than on the other.

When a verdict has been rendered against the weight of the evidence, the court may, on this ground, grant a new trial, but the court will exercise this power not merely with a cautious, but a strict and sure judgment, before they send the case to a second jury.

The general rule under such circumstances is, that the verdict once found shall stand: the setting aside is the exception, and ought to be an exception, of rare and almost singular occurrence. A new trial will be granted on this ground for either party; the evidence, however, is not to be weighed in golden scales.
   

This entry contains material from Bouvier's Legal Dictionary, a work published in the 1850's.