6. Simulate a measurement model

While it is becoming increasingly common to use numerical simulation in models of interest, it is less common to encounter this practice in measurement models. However, the simulation of an uncertainty budget enables us to verify the presumed normality of the mixture of all the factors contributing to the measurement. Indeed, the central limit theorem states that a mixture of laws tends towards a normal distribution, but it specifies that this is true if all the factors are independent and if none of them weighs, in variance, more than 30% of the total variance.

In the uncertainty budgets of measurement models, it is not uncommon for one cause to dominate over the others. It is also possible that certain factors contributing to uncertainty are correlated, which can disturb the normality of the resulting measurement uncertainty (cf. Bibliography "Covariance evaluation...