Fall 2026 bootcamps are sponsored by Aegon Transamerica
The Department of Statistics and Actuarial Science is excited to announce this year’s bootcamp schedule. Our distinguished presenters include Duncan Leaf, Congrui Yi, Jin Meng, and Kate Ralston, who will present on data analytics programming, business communication, and observational data analysis.
Review the 2026 schedule and learn more about the presenters
Dr. Duncan E. Leaf
Research Scientist, Center for Health Policy & Economics, University of Southern California
Dr. Leaf will give four presentations:
Understanding an Observational Data-Generating Process
This lecture sets up the motivation for understanding observational data and how it is different from randomized experimental data. While randomized experiments and observational studies are usually treated as two separate areas of statistics, I will argue that data from randomized experiments may be observational until proven otherwise. Therefore, understanding bias and causal reasoning is also useful when working with experimental data. I will introduce our working example of analyzing diabetes in the NHANES data set.
Complex Survey Sample Design and Analysis
This lecture walks through details of a complex survey design using NHANES as an example. We then look at weighting methods used to make a survey sample look like the population and variance estimation methods used to get valid inferences from survey samples, with R and SAS examples.
Bias and Causal Inference
The primary goal of this lecture is understanding the fundamental problems of causal inference. Counterfactual reasoning is introduced using potential outcomes notation. The secondary goal is to introduce basic causal analysis methods: inverse probability of treatment weighting and instrumental variable regression. R examples will be provided.
Project Presentations
Students will give two presentations of their NHANES data analysis project. The first presentation is non-technical and aimed at decision makers who have no knowledge or interest of statistical methods. The second presentation will explain your methods and results for a technical audience.
Dr. Congrui Yi
Machine Learning Engineer, Pinterest Labs
Dr. Yi will give two presentations:
Introduction to Multi-Armed Bandits
Multi-armed bandit (MAB) is a simple but powerful modeling framework for making sequential decisions given distributional uncertainty and dynamic environments. Given the right to choose an action at a time and collect corresponding rewards, MAB algorithms aim to maximize cumulative rewards by balancing two competing objectives: exploration of new or less chosen actions, and exploitation of ones shown to be rewarding. This lecture introduces basic concepts including action, reward, regret, exploration-exploitation trade-off, and most common types of MAB algorithms, such as epsilon-greedy, Upper Confidence Bound (UCB), Thompson Sampling (TS), and Boltzmann Exploration. Additionally, it will include a hands-on Python coding exercise and a discussion of real-world applications.
Introduction to Contextual Bandits
Following the introduction to MAB in lecture 1, this lecture continues to cover contextual bandits, which extends the classic MAB by incorporating contextual information, allowing us to handle more complex real world problems and make more informed decisions. We will go through several types of models including LinUCB (Linear Upper Confidence Bound), LinTS (Linear Thompson Sampling), BLIP-TS (Bayesian Linear Probit Regression with TS), Neural Contextual Bandits etc. Moreover, we will associate each type with examples of industrial applications in areas such as online advertising, recommender systems and dynamic pricing, with a focus on motivations and formulations.
Dr. Jin Meng
Senior Data Scientist, Zurich North America
Dr. Meng will give two presentations:
Data Science Application in Insurance on Traditional Tabular Data
In the first lecture, we will cover the relatively traditional approach of analyzing tabular data in the insurance field. The overall objective of this lecture is to give students both conceptual understanding and practical experience with data science workflows in a real-world insurance context.
This presentation will be held in person at 14 Schaeffer Hall (SH).
Data Science Application in Insurance on Unstructured Textual Data
In the second lecture, we will explore more cutting-edge techniques in natural language processing (NLP) using large language models (LLMs). The focus will be on building a chatbot solution, as a data science proof-of-concept solution utilizing a commonly used LLM and a Retrieval-Augmented Generation (RAG) framework to provide accurate answers to user questions based on external domain knowledge documents provided to the LLM. The goal of this lecture is to expose students to data science technique stack used in developing end-to-end proof-of-concept or prototype solutions based on cutting edge techniques.
This presentation will be held in person at 14 Schaeffer Hall (SH).
Dr. Kate Ralston
Director of Data Analytics, Enrollment Management, University of Iowa
Dr. Ralston will give two presentations:
Before You Build: Scope, Align, and Communicate It
The goal of this session is to build the communication and planning skills needed to successfully launch and manage data projects. Participants will learn to distinguish between business needs, project goals, and answerable data questions, gather and clarify stakeholder requirements, and develop realistic project scopes, timelines, and communication plans. Through discussion and practice, participants will consider how project complexity, uncertainty, timing, and stakeholder needs shape both technical choices and communication throughout a project. The session will also examine common setbacks, practical safeguards, and easy wins. Participants will leave with strategies for setting expectations, communicating proactively, and keeping data projects aligned with stakeholder needs.
This presentation will be held in person at 14 Schaeffer Hall (SH).
A homework will be assigned between these two sessions.
Making the Message Matter: Presenting Results with Purpose
The goal of this session is to help participants communicate data and analytical results in ways that are relevant, accessible, and actionable for different audiences. Participants will learn to identify stakeholder needs and tailor the content, level of detail, and format of a message to the audience. Through examples and practice, participants will learn to frame project objectives, highlight key findings, communicate implications, and determine when technical detail adds value. Participants will practice communicating results through oral and written formats, including an elevator pitch and executive summary. The session will also develop strategies for responding effectively to questions, critical feedback, and challenging conversations.
This presentation will be held in person at 14 Schaeffer Hall (SH).
Past bootcamps
2025
The Department of Statistics and Actuarial Science was excited to announce the 2025 bootcamp schedule. Our distinguished presenters included Duncan Leaf, Congrui Yi, Jin Meng, and Kate Ralston, who presented on data analytics programming, business communication, and observational data analysis.
Duncan Leaf, Research Scientist, Center for Health Policy & Economics, University of Southern California, gave four presentations:
- Understanding an observational data-generating process
- Complex survey sample design and analysis
- Bias and causal inference
- Project presentations
Congrui Yi, Senior Research Scientist, Meta, gave two presentations:
- Introduction to multi-armed bandits
- Introduction to contextual bandits
Jin Meng, Senior Data Scientist, Zurich North America, gave two presentations:
- Data science application in insurance on traditional tabular data
- Data science application in insurance on unstructured textual data
Kate Ralston, Director of Data Analytics, Enrollment Management, University of Iowa, gave two presentations:
- Before you build: How to scope, ask, and align like a pro
- Making the message matter: Presenting results with purpose
2024
Observational studies: Design and analysis
We are excited to announce the data science bootcamp offered by the Department of Statistics and Actuarial Science: "Observational Studies: Design and Analysis." This bootcamp is open to all Data Science, Statistics, or Actuarial Science students.
On Oct. 8th our speaker is Dr. Duncan E. Leaf, who is a research scientist at the USC Schaeffer Center for Health Policy & Economics. Learn more about Duncan.
Dr. Leaf's research centers on understanding the economic impacts of health policy decisions, with a particular focus on exploring the effects of palliative care interventions on medical costs and quality of life as well as simulating the long-term effects of early-childhood education interventions. In his BootCamp talk, Dr. Leaf will share key insights on effectively communicating complex health policy analysis, translating technical research into actionable strategies for policymakers, and bridging the gap between statisticians, health economists, and decision-makers to drive impactful change. His vast experience at the intersection of Statistics, Health Policy, and Economics makes him an invaluable speaker for anyone interested in the critical role of analytics in shaping economic policy.
Overview
Observational Studies: Design and Analysis
Duration: 8 hours
Dates: Tuesdays, Oct. 8, 15, 22, and 29
Time: 6 to 8 p.m. CST
Objectives: Students will learn key concepts, including:
- Introduction to observational data
- Complex survey sample design and analysis
- Addressing bias in observational studies
- Causal inference methodologies
The bootcamp will feature exercises and group discussions to reinforce these concepts.
Business communications boot camp - Part II
We are pleased to announce that Paul Hampton, Head of the Strategic Insights, Data, & Analytics team at Transamerica, will be our speaker for the upcoming Business Communications BootCamp - Part II. Learn more about Paul.
Paul leads a team dedicated to developing cutting-edge data infrastructure and analytics solutions that play a pivotal role in shaping business decisions at Transamerica. In his second boot camp talk, he will continue from the first part and share valuable insights on managing and expanding a data science team, effective strategies for communicating technical results to business leaders, and fostering collaboration between data scientists and decision-makers.
Paul will also discuss how clear and impactful communication of his team’s findings has contributed to both team growth and the overall success of the business.