North Carolina State University, Statistics Depart
Ana-Maria Staicu joined the Department of Statistics, North Carolina State University in 2009, after completing her Ph.D. with Nancy Reid and her postdoc at the University of Bristol, UK. Her research interests are primarily in functional data analysis, longitudinal data analysis, nonparametric statistics, and brain imaging analysis. It has been applied to brain tractography studies of MS and brain imaging studies more general, wearable computing, animal studies, environmental studies, social media, and others.
Significance Testing for Functional Effect in Crossover Design
Wearable devices for continuous monitoring of electronic health have increased attention due to their richness in information. Oftentimes inference is drawn from summaries that quantify some feature of the data, leading to a loss of information that could be useful when one utilizes the functional nature of the response. This work is motivated by the interest to understand the efficacy of meloxicam treatment for Osteoarthritis in household cats, by employing a crossover experiment and using the cats’ minute-by-minute activity levels as a proxy objective measure of the cat’s condition. We develop a testing procedure to test for the significance of the treatment effect in a crossover design, in the presence of the carryover effect. We propose an orthogonal projection-based test pseudo generalized F test and study its null asymptotic distribution under mild conditions. In numerical studies, the proposed test maintains the size, is powerful to detect the significance of functional treatment effect, and is very efficient compared to bootstrap-based alternatives.