Level: Intermediate
【Presenter】
Bill Worley, JMP Senior Global Enablement Engineer, SAS Institute Inc.
[Announcement summary]
In recent years, PLS regression (Partial Least Squares regression) has been used to create predictive models using spectral data. In this presentation, we will show an example of creating a good prediction model with less spectral data using a new approach that uses the covariates feature of Function Data Explorer (FDE) and Design of Experiments (DOE). This new method uses 1/4 to 1/3 of the data traditionally used to create predictive models. We will also introduce a model-based multivariate control chart (MDMCC), a new platform for JMP 15 as a method for analyzing spectral data.
[Presenter Profile]
Bill Worley had been with Procter & Gamble for 14 years before joining the JMP team, and was in a position to educate and support the use of JMP as a technology leader in the modeling / simulation group for the last five years. Worley is a chemist and holds a master's degree in chemistry from the University of Cincinnati.
