Presented on April 28, 2021- Invited presentation at an internal FDA Data Science workshop

When You do Applied Statistics, You're Acting Like a Scientist. Why Does this matter? by Andrew Gelman Visit https://rstats.ai/nyr/ to learn more. Abstract: When you do applied statistics, you form hypotheses, gather data, run experiments, modify your theories, etc. Here, I'm not talking about hypotheses of the form "theta = 0" or whatever; I'm talking about hypotheses such as, "N=200 will be enough for this study" or "Instrumental variables should work on this problem" or "We can safely use the normal approximation here" or "We really need to include a measurement-error model here" or "The research question of interest is unanswerable from the data we have here; what we really need to do is . . .", etc. Existing treatments of statistical practice and workflow (including in my own textbooks) do not really capture the way that the steps of statistical design, data collection, analysis, and decision making feel like science. We discuss the implications of this perspective and how it can make us better statisticians and data scientists. Bio: Andrew Gelman is a professor of statistics and political science and director of the Applied Statistics Center at Columbia University. He has received the Outstanding Statistical Application award from the American Statistical Association, the award for best article published in the American Political Science Review, and the Council of Presidents of Statistical Societies award for outstanding contributions by a person under the age of 40. His books include Bayesian Data Analysis (with John Carlin, Hal Stern, David Dunson, Aki Vehtari, and Don Rubin), Teaching Statistics: A Bag of Tricks (with Deb Nolan), Data Analysis Using Regression and Multilevel/Hierarchical Models (with Jennifer Hill), Red State, Blue State, Rich State, Poor State: Why Americans Vote the Way They Do (with David Park, Boris Shor, and Jeronimo Cortina), and A Quantitative Tour of the Social Sciences (co-edited with Jeronimo Cortina). Andrew has done research on a wide range of topics, including: why it is rational to vote; why campaign polls are so variable when elections are so predictable; why redistricting is good for democracy; reversals of death sentences; police stops in New York City, the statistical challenges of estimating small effects; the probability that your vote will be decisive; seats and votes in Congress; social network structure; arsenic in Bangladesh; radon in your basement; toxicology; medical imaging; and methods in surveys, experimental design, statistical inference, computation, and graphics. Twitter: https://twitter.com/StatModeling Presented at the 2022 New York R Conference (June 9, 2022)

C I Can Do More by Joe Marlo Visit https://d4con.io to learn more. Abstract: GitHub Actions for data science pipelines. Leveraging CI/CD for more than unit tests. Deploy apps, fit models, and score your data automatically. Bio: Joe is a Senior Data Scientist focused on statistics, Shiny, and R programming. He's authored multiple R packages on sequence analysis and causal inference diagnostics. Previously, he was a quantitative strategist at J.P. Morgan and a data scientist at Verizon. The former role leveraged proprietary big data to understand the spending and saving patterns of Americans and the latter role hierarchical forecasting problems. He holds a master's in applied statistics from New York University and a bachelor's in finance from University of Missouri. Joe is also an assistant research scientist at New York University where he works on thinkCausal, a project to make machine learning methods for causal inference more accessible. Twitter: https://twitter.com/LanderAnalytics Presented at the 2023 D4 Conference (August 24, 2023)

Mapping Big Data by Jared Lander Visit https://rstats.ai/gov to learn more. Abstract: Maps are one of the best forms of data visualization that readily understood while conveying a considerable amount of information. With the modern web, interactive, pannable, zoomable maps---known as slippy maps---have become the norm. Thanks to packages like {leaflet} it has never been easier to generate these maps. However, they don't scale well out of the box. We'll look at different methods for dealing with large data to make high performance maps. Bio: Jared P. Lander is Chief Data Scientist of Lander Analytics, the Organizer of the New York Open Statistical Programming Meetup and the New York R Conference and an Adjunct Professor of Statistics at Columbia University. With a masters from Columbia University in statistics and a bachelors from Muhlenberg College in mathematics, he has experience in both academic research and industry. Jared oversees the long-term direction of the company and acts as Lead Data Scientist, researching the best strategy, models and algorithms for modern data needs. This is in addition to his client-facing consulting and training. He specializes in data management, multilevel models, machine learning, generalized linear models, data management, visualization and statistical computing. He is the author of R for Everyone, the best-selling book about R Programming geared toward Data Scientists and Non-Statisticians alike. The book is available from Amazon, Barnes & Noble and InformIT. The material is drawn from the classes he teaches at Columbia and is incorporated into his corporate training. Very active in the data community, Jared is a frequent speaker at conferences, universities and meetups around the world. His writings on statistics can be found at jaredlander.com. Twitter: https://twitter.com/ Presented at the 2023 Government & Public Sector R Conference (October 19, 2023)

The Many Models in Production at NFL Next Gen Stats by Mike Band Visit https://rstats.ai/nyr to learn more. Abstract: Since its inception in 2016, the NFL's Next Gen Stats group has revolutionized football statistics. Through the utilization of player tracking data, NGS has developed a series of innovative metrics, many of which powered by distinct machine learning models. Each model delves into a unique facet of the game, contributing to comprehensive metrics that can evaluate the performance of not only individual players but entire teams and beyond. From Completion Probability to Expected Rushing Yards and the intuitive Fourth Down Decision Guide, I'll guide you on a fascinating journey through the many machine learning models in production at Next Gen Stats. Bio: Mike Band is a data scientist architect at Lander Analytics and an analyst with the National Football League’s Next Gen Stats team. With a master’s degree in analytics from the University of Chicago, and bachelor’s degree from the University of Florida, Mike specializes in driving actionability through data science projects. Working both as a data scientist and on the business development side of Lander Analytics, Mike’s combination of programming skills, applied statistics background and communication abilities help drive data science projects from start to finish. Mike’s key to creating value through analytics: Effective communication of complex results into comprehensible, need-to-know reporting personalized for the intended audience. Twitter: https://twitter.com/MBandNFL Presented at the 2023 New York R Conference (July 13, 2023)

Machine Learning for People by Gus Lipkin Visit https://d4con.io to learn more. Abstract: Through a practical example, we explore the strategic art of selecting machine learning models tailored to specific situational needs, shifting the focus beyond pure accuracy. We take into account both computational expense and ease of understanding when sharing models with others to build a balanced approach to model selection that combines both functional effectiveness and interpretability. Bio: Gus Lipkin is a Data Scientist at Lander Analytics. Prior to joining the team, he spent a year as a Data Analyst at Publix Super Markets where he focused on creating fully automated reporting pipelines, linking disparate data, and creating an R package to help optimize product mixes for any group of items. Gus earned his Bachelor's of Science in Business Analytics with concentrations in Quantitative Economics & Econometrics and Intelligent Mobility from Florida Polytechnic University where his Capstone project team won the Showcase Award. He lives in Florida with his two cats, enjoys spending time outdoors when it's not too hot, and plays Bananagrams and eats lots of coffee ice cream when it is. Twitter: https://twitter.com/guslipkin Presented at the 2023 D4 Conference (August 25, 2023)

Model Shootout: Comparing Linear Models, Trees and Neural Networks for Binary Classification by Jared P. Lander Visit https://d4con.io to learn more. Abstract: When analyzing data there are so many different kinds of models to choose from. Generalized linear models, tree-based models, neural networks, ensembles of all these. For a given problem set, we'll compare a number of differrent models and see which did best, to find the pareto optimum of performance and speed. Bio: Jared P. Lander is Chief Data Scientist of Lander Analytics, the Organizer of the New York Open Statistical Programming Meetup and the New York R and Government & Public Sector R Conferences, an Adjunct Professor at Columbia Business School, and a Visiting Lecturer at Princeton University. With a masters from Columbia University in statistics and a bachelors from Muhlenberg College in mathematics, he has experience in both academic research and industry. Jared oversees the long-term direction of the company and acts as Lead Data Scientist, researching the best strategy, models and algorithms for modern data needs. This is in addition to his client-facing consulting and training. He specializes in data management, multilevel models, machine learning, generalized linear models, visualization and statistical computing. He is the author of R for Everyone (now in its second edition), a book about R Programming geared toward Data Scientists and Non-Statisticians alike. The book is available from Amazon, Barnes & Noble and InformIT. The material is drawn from the classes he teaches at Columbia and is incorporated into his corporate training. Very active in the data community, Jared is a frequent speaker at conferences, universities and meetups around the world. He's an R Consortium Board Member. His writings on statistics can be found at jaredlander.com. Twitter: https://twitter.com/jaredlander Presented at the 2023 D4 Conference (August 24, 2023)

Your Maps Might Be Lying to You by Gus Lipkin Visit https://rstats.ai/gov to learn more. Abstract: Bio: Twitter: https://twitter.com/ Presented at the 2023 Government & Public Sector R Conference (October 20, 2023)