The t-Distribution Approach to the Second-Order Multiresponse Surface Model of Paracetamol Tablets Quality Characteristics

The normal distribution approach is often used in regression analysis at the Response Surface Methodology (RSM) modeling stage. Several studies have shown that has drawbacks compared to more robust t-distribution approach. found control size much successfully small samples existing methods presence of moderately heavy tails. In many RSM applications, there than one response (multiresponse), which usually correlated with each other (multivariate). On hand, actual surface habitually indicates curve by optimal value, so second-order model used. This paper aims develop multiresponse using a multivariate work also provides parameter estimation procedure and hypothesis testing for significance parameter. First, performed Maximum Likelihood Estimation (MLE), followed Expectation–Maximization algorithm as an iterative method find (local) maximum likelihood. Next, Ratio Test (LRT) test parameters simultaneously. obtained uses this determine conditions input variables optimize Paracetamol tablets’ physical quality characteristics.