This study focuses on the development of a generalised multivariate beta type II distribution as well as the noncentral and bimatrix counterparts with positive domain. These models emanate from a sequential quality monitoring procedure with the normal and multivariate normal distributions as the underlying process distributions. Three different scenarios are considered, namely: 1. The variance is monitored from a normal process and the mean remains unchanged; 2. The above-mentioned scenario but the known mean also encounters a sustained shift; 3. The covariance structure of a multivariate normal distribution is monitored with the known mean vector unchanged. The statistics originating from the above-mentioned scenarios considered are constructed from different dependent chi-squared or Wishart ratios. Exact expressions are derived for the probability density functions of these statistics. These new distributions contribute to the statistical discipline in the sense that it can serve as alternatives to existing probability models, and can be used in determining the performance of the quality monitoring procedure. / Thesis (PhD)--University of Pretoria, 2013. / gm2014 / Statistics / unrestricted