BLACK FRIDAY SALE IS NOW LIVE!!!

Advanced Bayesian Modeling in Stan

“Scott's approach to teaching Bayesian statistics is a masterclass in clarity and depth. With his unique ability to blend statistical modeling with real-world applications like sports analytics, he unravels the complexities of RStan coding and Bayesian modeling. From understanding the intricacies of physics phenomena to delivering captivating narratives through detailed visualizations, Scott is the ideal guide to jumpstart your journey into Bayesian modeling. ”

Michael S. Czahor, PhD

President @ Athlyticz

Releasing January 2025

Preorder our Pro-level Bayesian training and learn from Columbia Professor, Dr. Scott Spencer.

becoming a bayezian II course

Course Information

  • AthlyticZ Pro Series, an Immersive, masterclass style coding framework with Dr. Scott Spencer

  • 100+ Lessons (20+) Modules of engaging content

  • Asynchronous, self-paced learning

  • Pre-recorded video lessons for flexible scheduling

  • Interactive Features: Quizzes throughout the course to reinforce learning

  • Supplementary Resources: Access to downloadable slides, readings, and additional materials

  • Hands-On Projects: Dive into detailed case studies with Dr. Scott Spencer.

  • Virtual Machines: Immerse yourself in our customized IDE solutions for real world programming experience.

$1199 usd

Your Bayesian Analysis Coach for Becoming a BayeZian II

My passion lies in leveraging data for good causes and exploring the intricate dynamics of professional sports through statistical modeling.

Scott spencer

Dive into the world of Bayesian analysis and cutting-edge probabilistic programming alongside a Columbia Professor and Stan language collaborator with expertise in crafting intricate generative models. Scott's expertise spans from decoding human behavior to forecasting sea-level rise impacts on coastal property values, all while dissecting the statistical DNA of professional sports.

Scott's influence extends beyond academia, shaping decisions for tech giants like Amazon, healthcare leaders such as Johnson & Johnson, and entertainment moguls like Vevo. His knack for transparent storytelling through R packages ensures that even the most complex insights are accessible and actionable, making him a sought-after guide for data enthusiasts across industries.

Join the BayeZian revolution and unlock the true potential of Bayesian methods with Scott, where every analysis tells a compelling story and uncertainty is the key to innovation!"

What You'll Learn

Introduction and Workflow

  • Intro: High-level review of Becoming a Bayesian I

  • Intro: Roadmap

  • Workflow: Before fitting a model

  • Workflow: Fitting a model

  • Workflow: Simulating data

  • Workflow: Addressing computational problems

  • Workflow: Evaluating and using a fitted model

  • Workflow: Modifying a model

  • Workflow: Understanding and comparing multiple models

    • Model selection

    • Model averaging

Mixture, Rating, & Ranking Models +Advanced Multilevel Models

  • Mixture Models: Overdispersion

  • Mixture Models: Exploring baseball scores as Poisson distributed

  • Mixture Models: Negative binomial mixture of Poisson distributions

  • Mixture Models: Baseball scores as mixture of Poisson distribution

  • Mixture Models: Zero-inflation processes and hurdle models

  • Rating & Ranking Models: Historical overview as a family

  • Rating & Ranking Models: An overview of the family

    • Ordinal Regression

    • Rank-ordered logistic regression

    • Bradley-Terry models

    • Plackett-Luce models

    • Rating Models (Elo, Glicko)

    • Unifying principles of ordered and rank models

  • Mixture Models: Ordinal regression

    • General understanding of ordinal regression

    • Example code in R and Stan

    • Example draft order and early-career performance

  • Mixture Models: Rank-ordered logistic regression

  • Advanced Multilevel Models: Common approach omits information

  • Advanced Multilevel Models: Multilevel structure propagates the uncertainty

  • Advanced Multilevel Models: Applied examples

Sufficient Statistics, Advanced Correlation, QR Parameterization, & Autoregressive

  • Sufficient Statistics

  • Advanced Correlation: Trivariate reduction

  • Advanced Correlation: Marginal plus conditional

  • Advanced Correlation: Copulas

  • QR Parameterization

  • Autoregressive: AR processes with equal times between measures

  • Autoregressive: AR processes with irregular times (gaps) between measures

  • Multiple AR processes with interactions

Survival Analysis

  • Survival Analysis: Time-to-events, proportional to power law

  • Survival Analysis: Simulate time-to-even data

  • Survival Analysis: General hazard and survival functions

  • Survival Analysis: Weibull survival model without covariates

  • Survival Analysis: Weibull model with covariates

  • Survival Analysis: Modifying the priors for baseball

  • Survival Analysis: Fitting a model to baseball data

  • Survival Analysis: Interpreting the coefficients of the covariates

  • Survival Analysis: Discretizing the Weibull Distribution

  • Survival Analysis: Discrete Hazard Function

  • Survival Analysis: Log-Hazard in discrete time

  • Survival Analysis: Discrete-time Stan model

  • Survival Analysis: Estimate parameters from baseball data

  • Survival Analysis: Posterior predictive checks

  • Survival Analysis: Posterior inference

Differential and Difference Equations

  • Differential Equations

  • Difference Equations: Modeling performance with Bannister's Impulse-Response Model (Example)

  • Difference Equations: Cycling power data for training (Example)

  • Difference Equations: Implementing the model in Stan

  • Difference Equations: Prep data

  • Difference Equations: Compile and fit the model

  • Difference Equations: Review the estimates

Splines

  • Splines: Model overview

  • Splines: B-Spline construction for a single variable

    • Construction in Stan

    • Construction in R

  • Splines: Regression on 𝐵(𝑥)

  • Splines: Tensor product of spline bases

    • Tensor product of vectors

    • Tensor product of functions

    • Tensor product of matrices

  • Splines: Two-dimensional splines

  • Splines: Implementing th

  • Splines: Regression on 𝐵(𝑥) ⊗ 𝐵 (𝑦)

  • Splines: Prediction for new data

  • Splines: Model comparison

  • Splines: Implications

Gaussian Processes (GPs)

  • GPs: Likelihood

  • GPs: Gaussian process for latent 𝑓

  • GPs: Understanding the Cholesky decomposition in Gaussian process

  • GPs: Using the Cholesky decomposition

  • GPs: Hyperparameters and priors

  • GPs: Estimates at new or counterfactual 𝑥

    • Covariance for new points

    • Conditional mean for 𝑓new

    • Conditional covariance for uncertainty

    • Generating predictions

  • GPs: Stan code for N-Dimensional Gaussian process

  • GPs: Testing the code in one dimension

  • GPs: Testing the code in two dimensions

Hilbert-Space Approximate Gaussian Processes (HSGPs)

  • HSGPs: Math refresher, Fourier transforms

  • HSGPs: Basis functions generally

  • HSGPs: Frequencies in basis functions

  • HSGPs: More math, spectral densities

  • HSGPs: Basis functions 𝜙

  • HSGPs: Weighted basis function sum to model 𝑓(𝑥)

  • HSGPs: Likelihood for observed data

  • HSGPs: Priors for parameters

  • HSGPs: Full model specification

  • HSGPs: Visual walkthrough using R

  • HSGPs: Implementation of a N-dimensional HSGP in Stan

  • HSGPs: Using the model

Physics-Constrained Models, Common Issues, Computational Performance, and Next Steps

  • Physics-constrained models

    • Applied Example I

    • Applied Example II

    • Applied Example III

    • Applied Example IV

  • Common Issues

    • Outliers and robustness

    • Missing data imputation

    • Censoring and truncation

    • Parameter-space transformations

  • Computational Performance

    • Coding optimizations

    • GPU

    • Within-chain parallelization

    • Memory

  • Next Steps

master Bayesian analysis with Becoming a BayeZian II

How We Conduct Our Course

At AthlyticZ, we've designed our course structure to empower your learning journey at your own pace, ensuring flexibility and accessibility.

Prerecorded Video Lessons

Delve into our comprehensive course content with prerecorded video lessons, allowing you to learn at your convenience and revisit concepts as needed.

Assignments and Exercises

Engage with assignments and exercises that reinforce your understanding of key concepts and help you build practical skills.

Continuous Assessment

Benefit from quizzes and assessments integrated throughout the course, providing you with ongoing feedback on your progress and understanding.

Hands-On Projects

Apply your knowledge to real-world projects, including both minor and major assignments, where you'll develop comprehensive models using Stan.

Supplementary Resources

Access a wealth of supplementary resources such as readings, downloadable slides, and personalized tutorials tailored to your interests, enriching your learning experience.

Interactive Features

Utilize interactive features within each lesson, including note-taking capabilities and custom-built IDEs with preloaded code, ensuring an immersive and engaging learning environment.

MASTER advanced Bayesian analysis skills with Becoming a BayeZian II

"Transformative Learning Experience"

"AthlyticZ has completely transformed the learning approach to data science through the use of sports-based problems. The course structure is intuitive, the content is comprehensive, the instructors are the best of the best, and the practical projects have immediate impact to students."

BILL GEIVETT M.ED.

PRESIDENT @ IMA TEAM

AUTHOR: DO YOU WANT TO WORK IN BASEBALL?

SR. VP: COLORADO ROCKIES 2001-2014

ASST. GM: LOS ANGELES DODGERS 1998-2000

Frequently Asked Questions

Can I expense this with my employer?

Unlock the transformative potential of Bayesian analysis with our course. Here's how your employer can harness the power of Becoming a BayeZian:

  • Streamline Data Analysis: Arm your team with advanced Bayesian skills to streamline data analysis processes, leading to quicker insights and informed decision-making.

  • Optimize Workflows: Develop tailored post-processing workflows aligned with your company's requirements, enhancing efficiency and minimizing manual data manipulation tasks.

  • Cost-Effective Solutions: Reduce expenses on commercial software by utilizing open-source Bayesian analysis tools and R packages for comprehensive data analysis and visualization.

  • Empower your team to excel in data analysis and propel business growth with our comprehensive Becoming a BayeZian course.

Do I need access to any commercial software?

No, you won't need any commercial software for this course. All Stan models will run through our prepackaged VMs built for your convenience.

Does the course assume extensive knowledge of R ?

We recommend having intermediate Tidyverse R skills for this course. Our BreeZing through the Tidyverse Course is a great choice for prerequisite knowledge in R. Additionally we recommend an understand of course material from Becoming a BayeZian I.

Is the course graded? What defines successful completion?

Evaluation is based on the completion of course work rather than "correct answers." Successful completion entails:

• 80% or higher on all quizzes

• 100% of modules completed

Upon meeting these criteria, you'll receive a certificate of completion.

What if I need to drop out? Is there a refund policy?

At AthlyticZ, we're committed to your success and satisfaction throughout your learning journey.

Our refund policy is as follows:

  • You can drop the course within 3 days of commencement for a full refund. After this date, refunds are not provided.

  • No refunds will be granted if you have completed 25% or more of the course material, regardless of the time elapsed since the course began.

    Thank you for choosing AthlyticZ for your educational needs.

© 2024 AthlyticZ

Designed With ❤️ By Jackson Yew

Privacy Policy · Terms of Use · Terms of Sale ·  Email questions to : [email protected]