2025 FTC Short Courses

Short courses will be held on October 7, 2025. Attendees may registers for the conference and short courses together or separately. Conference attendance is not required to attend a short course.

Full Day

Annie Booth  – High Quality Coding for High Quality Analytics

Half Day

Jon Stallrich – A Practitioner’s Guide to Optimal Design

Mindy Hotchkiss – Practical Experimentation: Improving the Logistics of Experiment Management

FULL DAY COURSE

8:00am to 5:00pm (1 hour lunch at 12:00pm) 

High Quality Coding for High Quality Analytics

Annie Booth, Virginia Tech

Abstract:  Across research and industry, we all use statistical programming languages to simulate, synthesize, visualize, model, and understand data.  But what makes for high quality code?  When we think of quality we think of efficient processes with little or no waste which are consistent, reliable, and robust.  The same principles should apply to our codes.

This two-part course will focus on strategies for coding efficiently and effectively.  A particular emphasis will be on AUTOMATING things that are often repetitive (such as generating figures and summaries of updated data).  The morning and afternoon sessions will focus on the R and python programming languages, respectively.

For both languages, topics will include:

  • Using class objects and functions. These are great for automating things you tend to do over-and-over again with similar data.  No more copying and pasting chunks of code!
  • Wrapping your codes inside packages. Many of us are familiar with formally published packages from CRAN/pip but fail to leverage these user-friendly formats in our own work. Wrapping your codes in a package creates an easier, streamlined interface.  It is worthwhile even if you’re the only one to use it!
  • Using command line arguments to repeatedly run codes under different settings. This is a huge time saver if you are running similar jobs over-and-over again.  As an added bonus, it allows for easy parallelization.

Target audience and prerequisites:

CONTACT: annie_booth@vt.edu

Annie Booth is an Assistant Professor in the Department of Statistics at Virginia Tech, where she earned her Ph.D. in Statistics in 2023. Before returning to Virginia Tech as a faculty member, she was an Assistant Professor of Statistics at NC State University. Her Ph.D. dissertation was recently selected as a finalist for the 2023 ISBA Savage Award. She specializes in surrogate modeling of computer experiments including uncertainty quantification, active learning, Bayesian optimization, and reliability analysis. In her work she emphasizes reproducibility and accessibility through the publication of open-source easy-to-use software.

HALF DAY COURSES

A Practitioner’s Guide to Optimal Design
8:00am to 12:00pm

Jon Stallrich, North Carolina State University

Abstract: What makes one experimental design better than another? In this short course, I will first introduce a general optimal design framework that answers this question by ranking designs according to a so-called “design criterion” that summarizes the amount of statistical information held by a design. Popular design criteria, such as the D- and A-criterion, that seek to minimize estimation variances will be discussed and compared. This comparison will spark a discussion on the importance of choosing an appropriate design criterion to match your analysis goals, leading to an introduction to more flexible variance-based criteria and criteria to minimize estimator bias. Computational search algorithms will be presented to help identify optimal (or nearly optimal) designs for a general criterion. R code will be provided that perform these search algorithms. Examples will focus on one factor studies, factorial experiments, and response surface designs. By the end of the course, participants will be experts in the general optimal design framework; be familiar with popular optimality criteria targeting the variance and bias of least-squares estimators; and be able to program design search algorithms to optimize a general criterion.

Target audience and prerequisites:

CONTACT:jwstalli@ncsu.edu

Joh Stallrich Jon Stallrich is an Associate Professor in the Department of Statistics at North Carolina State University. He earned his Ph.D. in Statistics from Virginia Tech in 2014. His research interests include design and analysis of screening experiments, computer experiments, online controlled experiments, functional data analysis, and variable selection. He has served in multiple leadership roles for the American Statistical Association, including member of the Committee for Applied Statisticians, Program Chair and Chair for SPES, and General Conference Chair for the 2024 Fall Technical Conference. In 2021, his co-authored paper won the ASA SPES Award recognizing excellence in partnerships among statisticians, scientists, and engineers across the many disciplines encompassed by the physical and engineering sciences.

Practical Experimentation: Improving the Logistics of Experiment Management
1:00pm to 5:00pm

Mindy Hotchkiss, Enquery Research

Abstract:  The process of creating statistically structured designed experiments, known as the Design of Experiments (DOE/DOX), is widely recognized as best practice for test planning. It is used in countless fields and industries to facilitate knowledge development and establish causality, which is foundational to repeatable research. DOE is a multipurpose tool that uses balance and structure to effectively characterize, improve, and optimize systems and processes. However, good test practices are critical for obtaining robust results that can be reproduced by other experimenters, enabling extension into further research.

The objective of experimentation is, in its simplest form, attempting to detect a signal clouded by some degree of noise (e.g. natural process variability). For example, the most common form is to establish if a change in factor X is inducing an effect in some response Y. Statistical analysis methods are used to quantitatively differentiate the signal from the noise. However, there are any number of logistical reasons that could contribute to why a particular signal could not be detected, such as poor factor level selection, suboptimal blocking, high measurement variability in independent and dependent variables, unstable test setup, improper data post-processing, etc.. To improve the clarity of experimental results, there are two options: one must either increase the signal, decrease the noise, or some combination of the two.

Since experiments rarely if ever have an opportunity to be redone, decisions made during the test planning process have the potential for enormous downstream consequences for the outcome of the entire experiment. It is critical that experimenters consider what could contribute to noise during the test planning process, in order to most effectively achieve the benefits of the investment made in planned testing. This course will discuss practical test strategies to both enhance the signal and reduce noise, as well as general experimental practices that can be utilized during the experimental design and test planning process, as well as implementation. The objective of this course is to provide a resource for practitioners, those implementing experiments, both as statisticians and other subject-matter experts involved in projects as part of cross-functional teams.

Target audience and prerequisites:

CONTACT: 

Mindy Hotchkiss is an independent consultant operating as Enquery Research LLC. She was Technical Specialist and enterprise-wide Subject Matter Expert in Statistics at Aerojet Rocketdyne, now an L3Harris Technologies company. She has over 25 years of experience as a statistical consultant between Pratt & Whitney and Aerojet Rocketdyne, supporting engineering, operations, and technology development across the enterprise, including hypersonics and additive manufacturing, and where she spent many years facilitating complex multi-phase experiments in a wide variety of different physical and virtual environments. She has BS degrees in Mathematics and Statistics and an MBA from the University of Florida, and a Masters of Statistics from North Carolina State University. She is an ASQ CRE, CQE, and CSSBB, and a Past Chair of the ASQ Statistics Division.