For example, a singularity occurs in fitting statistical models to data when we include
confounded parameters in the model. This could happen if one collected data from an experiment conducted over several years at a different site in each year, and then tried to fit a model to simultaneously estimate the effects of years and sites. In linear regression or mixed models analyses, this would result in a singularity in a matrix that must be inverted in order to solve for the model effects, and no unique solution can be given.