Skip to content

Undergraduate Research with Graduate Topics & Experimental Design with Tidyverse

Photo of
Hosted By
Madeline and 4 others

Details

# Schedule
Tuesday, March 16, 6:30-8:00 pm (Pacific)
6:30 - Welcome and Networking
6:55 - Announcements
7:00 - Undergraduate Research with Graduate Level Topics
7:25 - Q&A
7:30 - Modernizing Design of Experiment Output with Tidyverse
7:55 - Q&A
8:00 - Meeting Adjourned

# Undergraduate Research with Graduate Level Topics
Speaker: Jessica Jaynes, PhD
Assistant Professor of Statistics, Department of Mathematics, CSU Fullerton

Bio: Dr. Jessica Jaynes is an Assistant Professor of Statistics in the Department of Mathematics at California State University, Fullerton. She received her Ph.D. in Statistics from the University of California, Los Angeles. Her area of expertise includes design of experiments, statistical methods and applications, survey methods, and discrete choice experiments. Her recent projects include the development of a new class of experimental designs that provide higher efficiency, more in-depth analyses, and require a shorter time to run. This research has led to multiple interdisciplinary funded projects.

Abstract: As science and technology progress, this often calls for innovation in methodological and theoretical development of statistical experimental design, an essential aspect for research in any discipline. This research concentrates on the close connection between statistics and medicine, and the development of novel experiments to maximize efficiency, while minimizing time and costs. The focus of this talk will be on an application of an undergraduate research project using a novel experimental design known as an Orthogonal Array Composite Design (OACD) with applications to KB oral cancer. In particular, this talk will highlight the methods and applications that undergraduate students contributed to. Furthermore, various takeaways will be presented in terms of tips, challenges, and benefits in working with undergraduate students. sources.

# Modernizing Design of Experiment Output with Tidyverse
Speaker: Jose Toledo
Graduate Student, Statistics Dept, CSU Fullerton

Bio: Jose is currently in his last semester in the Master’s in statistics program at California State University, Fullerton (CSUF), where he is also a teaching associate. Additionally, he is a part-time data analyst at Southern California Coast Water Project. His research interest is in the field of Design of Experiments, where he has worked on projects such as drug combination treatments for KB oral cancer and recently in tuberculosis. During his summer breaks he enjoys mentoring undergraduate students in local community colleges and providing them research opportunities early in their education.

Abstract: Most Design of Experiment (DoE) experts will agree that using FrF2 is the most comprehensive R package for generating two-level factorial designs as well as Plackett-Burman type screening designs. Similarly, for sequential optimization with first order and second order response surface models, that offer optimization approaches like steepest ascent and visualization of the response function for linear model objects, the package rsm is highly recommended. However, their graphical outputs are limited and can benefit from the current “trends” in R such as implementing graphics using ggplot2. This talk will focus on the development of an R package, ggDoE, which focuses on the visualization of DoE using ggplot2. Examples will be taken from a TB drug combination experiment and demonstrate the output of commonly used methods in DoE. These methods include but are not limited to: factorial effect plots, half-normal plots, Lenth’s method, fitting second-order models and their corresponding contour plots.

# Meeting Materials
All meeting materials will be sent to registered attendees after the event.

Online event
This event has passed