We consider the problem of parametric statistical inference when likelihood computations are prohibitively expensive but sampling from model is possible. Several so-called likelihood-free methods have been developed to perform in absence a function. The popular synthetic approach infers parameters by modelling summary statistics data Gaussian probability distribution. In another called approxim...

BACKGROUND Likelihood ratios offer important advantages over sensitivity and specificity for characterizing diagnostic tests. They can capture the magnitude of abnormality of test results, whereas sensitivity and specificity require that the test results be dichotomized into positive or negative. This is an important advantage because many diagnostic tests are measured on continuous or ordinal ...

The likelihood ratio approach for DNA typing in criminal cases is described. It is shown how this approach uses both the sizes and pattern of discrepancies between the crime scene profile of fragment lengths and the suspect profile for quantifying the strength of the evidence. In contrast to the current match-binning approach, it avoids an initial decision about whether the two profiles match. ...