ABSTRACT
The key point of difference of a continuous model is that it considers the peak heights as a continuous variable. This chapter describes the biological models and statistics behind STRmix™, one such continuous model. Such models may require some preprocessing, say of stutter peaks, or may be fully automated. These methods have the potential to handle any type of non-concordance and may assess any number of replicates without heuristic preprocessing and the consequent loss of information. The more correct and relevant information a system is able to make use of, the better its ability will be to differentiate true from false donors. The false exclusions for STRmix are usually caused by unusual polymerase chain reaction behaviour. STRmix uses standard mathematics and a model for peak and stutter height. The Markov chain Monte Carlo trials numerous combinations of parameter values to describe the observed data and ultimately generates posterior distributions for each parameter in the model.
