Do not confuse, here, the summation symbol X with matrix £.
The above equations are for the joint probability density of p independent Multi- unidimensional normal variables yj. It is easy to go from there to the multinormal normal distribution, where y is a p-dimensional random variable whose p dimensions are not independent. In order to do so, one simply replaces the above matrix X by a dispersion matrix with variances and covariances, i.e. (eq. 4.2):
Was this article helpful?