Functional data analysis and nonlinear regression models: an information quality perspective

Abstract

Data from measurements over time can be analyzed in different ways. In this article, we compare functional data analysis and nonlinear regression models using, among others, eight information quality dimensions. We present two case studies. The first case study introduces functional data analysis and nonlinear regression models in analyzing dissolution profiles of drug tablets where profiles of tablets under test are compared to reference tablets. A second case study involves statistically designed mixture experiments used in optimization tablet formulation. Python and JMP features are used to demonstrate the methods used in the two case studies.

Keywords:

• dissolution curve
• functional data analysis
• information quality
• mixture experiments
• nonlinear regression

Additional information

Notes on contributors

Ron S. Kenett

Professor Ron Kenett is Chairman of the KPA Group, Israel, Chairman of the Data Science Society at AEAI, Senior Research Fellow at the Samuel Neaman Institute, Technion, Haifa, Israel. and Research Professor at the University of Turin, Italy. He is an applied statistician combining expertise in academic, consulting and business domains. Ron is member of the Public Advisory Council for Statistics Israel, member of the of the executive academic council, Wingate academic college for sports education, member of the INFORMS QSR advisory board, member of the advisory board of DSRC, the University of Haifa Data Science Research Center and member of the board of directors in several startup companies. He authored and coauthored over 250 papers and 16 books on topics such as data science, industrial statistics, biostatistics, healthcare, customer surveys, multivariate quality control, risk management, system and software testing, and information quality. He was awarded the 2013 Greenfield Medal by the Royal Statistical Society and, in 2018, the Box Medal by the European Network for Business and Industrial Statistics.

Chris Gotwalt

Dr. Chris Gotwalt leads the statistical software development and testing teams for JMP Statistical Discovery. His passion is developing new technologies that accelerate innovation in industry and science. Since joining the company as a PhD student intern in 2001, Gotwalt has contributed many numerical algorithms and new statistical techniques. He has authored algorithms in JMP for fitting neural networks, linear mixed models, optimal design of experiments, analytical procedures for text analysis, and the algorithms for fitting structural equation models. Gotwalt is a principal investigator for Self-Validating Ensemble Models (SVEM), a procedure that makes machine learning possible for the small data sets often encountered in industry. He holds adjunct professorial positions at North Carolina State University, University of Nebraska and University of New Hampshire, and was the 2020 Chair of the Quality and Productivity Section of the American Statistical Association.