Hierarchical modeling for ‘zoom-in/zoom-out’ and multi-block data analytics

Tuesday, May 23, 2017

Time: 15.00-16.00 (CEST)

Sartorius Stedim Data Analytics invites to a webinar on hierarchical modeling in SIMCA 14.1

Data volumes increase for every day! Wherever you go there are data and you make data-driven decisions on a routine basis. Data analytics are needed to extract the information from the heaps of data. However, multivariate models involving large numbers of variables are often difficult to interpret, because plots become cluttered and difficult to overview. There may then be a temptation to eliminate variables which runs the risk of losing important information and may make the modeling efforts more difficult to interpret and less robust. 

The hierarchical approach to multivariate analysis greatly enhances our ability to penetrate complex data analytical problems. The zoom-in/zoom-out capability allows us first to understand complex relationships in terms of process or system blocks, and, then, to zoom-in on individual blocks to resolve the details in terms of the underlying process or system variables.

Topics for this webinar

  • An introduction to hierarchical modeling
  • Hierarchical modeling as an alternative to variable selection
  • Hierarchical modeling as alternative to block-scaling
  • Tools for ensuring the right level of detail is inspected
  • Drill-down capabilities

Click here to register