Join Prof. Frank E. Harrell Jr. (Vanderbilt University), a leading authority in applied biostatistics and regression methodology, and Drew G. Levy, PhD, for an online seminar designed for PhD students and professional researchers seeking to strengthen their quantitative research practice. Drawing on Regression Modeling Strategies (Harrell, 2015) and the continuously updated online materials, this workshop covers contemporary approaches to specifying, fitting, validating, and interpreting multivariable regression models for prediction and inference. Participants will explore key challenges in empirical research, including nonlinearity and interactions, overfitting, missing data, model validation, causal inference-based model specification, comparisons between regression and machine-learning approaches, and modeling continuous, binary, ordinal, longitudinal, and survival outcomes. With a strong emphasis on principled model formulation, graphical interpretation, reproducible workflows, and applied R examples using the rms package, the seminar equips participants to produce more robust, transparent, and publishable analyses. The seminar is available via live Zoom sessions or asynchronously with recordings and materials, includes access to an expert-moderated Q&A forum, and provides an official Instats certificate of completion. #BayesianStatistics #Biostatistics #CausalInference #ComputationalStatistics #DataScience #Epidemiology #HealthSciences #LongitudinalDataAnalysis #MachineLearning #MissingData #RegressionModelingStrategies #Statisticalmodellingandregression #Statistics #PublicHealth #R #Research #ResearchTraining #Instats

The hidden assumptions behind the data we observe.

Strengthen the credibility and impact of your empirical research with our online seminar on contemporary regression modeling strategies for PhD students and professional researchers. Drawing on Frank E. Harrell Jr.’s Regression Modeling Strategies and taught by Prof. Harrell, with causal inference-based model specification led by Drew G. Levy, PhD, this workshop covers principled approaches to specifying, fitting, validating, and interpreting multivariable regression models for prediction and inference. Participants will explore key challenges in quantitative research, including nonlinearity and interactions, overfitting, missing covariates, model validation, comparisons with machine learning, and modeling for continuous, binary, ordinal, longitudinal, and survival outcomes. Through focused instruction, applied case studies, and reproducible R-based examples using the rms package, attendees will learn practical strategies for spline modeling, shrinkage and penalization, bootstrap and cross-validation, causal-informed variable selection, and publication-ready visualization of complex model results. The seminar is available live via Zoom or asynchronously with recordings and materials, includes access to a monitored Q&A forum after the live sessions, and provides an official Instats certificate of completion. #BayesianStatistics #Biostatistics #CausalInference #ComputationalStatistics #DataScience #Epidemiology #HealthSciences #LongitudinalDataAnalysis #MachineLearning #MissingData #RegressionModelingStrategies #Statisticalmodellingandregression #Statistics #PublicHealth #R #Research #ResearchTraining #Instats

Strengthen your quantitative research practice with this online seminar on contemporary regression modeling strategies for PhD students and professional researchers, taught primarily by Prof. Frank E. Harrell Jr. of Vanderbilt University, with a dedicated component on causal inference-based model specification led by Drew G. Levy, PhD. Drawing on Regression Modeling Strategies and the updated online materials, the seminar covers principled approaches to specifying, fitting, validating, and interpreting multivariable regression models for prediction and inference across continuous, binary, ordinal, longitudinal, and survival outcomes. Participants will explore key challenges in empirical research, including nonlinearity and interactions, overfitting, missing or partially observed covariates, variable selection, shrinkage and penalization, Bayesian and semiparametric ordinal models, model validation using bootstrap and cross-validation, graphical interpretation, and comparisons between statistical regression and machine-learning approaches. With worked case studies, reproducible R examples, and guidance using the rms package, this seminar is designed to help researchers produce more robust, transparent, and publication-ready analyses. Live sessions are offered via Zoom, with recordings, materials, and an expert-monitored Q&A forum available for continued learning after the seminar. An official Instats certificate of completion is provided, with ECTS Equivalent points available where applicable. #BayesianStatistics #Biostatistics #CausalInference #ComputationalStatistics #DataScience #Epidemiology #HealthSciences #LongitudinalDataAnalysis #MachineLearning #MissingData #RegressionModelingStrategies #Statisticalmodellingandregression #Statistics #PublicHealth #R #Research #ResearchTraining #Instats

22-24 April 2026

22-24 April 2026

Advance your empirical research with an intensive online seminar on contemporary strategies for multivariable regression modeling for prediction and inference, led by Prof. Frank E. Harrell Jr. (Vanderbilt) with causal-inference sessions by Drew G. Levy, PhD. The course covers principled model formulation, handling nonlinearity and interactions (splines and smoothers), avoiding overfitting (shrinkage, penalization, resampling validation), missing and partially observed covariates, Bayesian and semiparametric ordinal approaches, and modeling for continuous, binary, longitudinal, and time-to-event outcomes—plus comparisons with machine-learning alternatives. Practical, reproducible workflows with R (rms package), worked case studies, publication-ready graphics, and R code are emphasized; prior competence with ordinary multiple linear regression is expected. Live-streamed via Zoom with recordings and materials available, a 30-day expert-monitored Q&A forum, on-demand access for 30 days after activation, and a certificate of completion (ECTS equivalent where indicated) are provided. Ideal for PhD students and professional researchers seeking to strengthen methodological skills and produce more credible, publishable analyses—register now to secure your place. #BayesianStatistics #Biostatistics #CausalInference #ComputationalStatistics #DataScience #Epidemiology #HealthSciences #LongitudinalDataAnalysis #MachineLearning #MissingData #RegressionModelingStrategies #Statisticalmodellingandregression #Statistics #PublicHealth #R #Research #ResearchTraining #Instats

A Note on Dropping Experimental Subjects who Fail a Manipulation Check - Volume 27 Issue 4

22-24 April 2026

👁️ What if your free will wasn't yours to keep? Think the Epstein saga ended in a Manhattan jail cell? Think again. Following the explosive Epstein Files Transparency Act document release of 2026, the world is finally seeing the redacted truth that the DOJ tried to bury. This isn't just a story of corruption; it's a deep dive into a chilling botanical mind control operation using Scopolamine—the infamous "Devil’s Breath"—to compromise the world's most powerful figures and their victims. In this episode, we peel back the layers of: - 🧪 The Pharmacology of Control: How Epstein used Angel’s Trumpet plants to facilitate memory erasure and chemical coercion. - 🏛️ The Great Fall of 2026: We detail the fallout of the Prince Andrew arrest and the sudden resignations of global power players like Casey Wasserman and Peter Mandelson. - 📹 The Surveillance State: New evidence regarding the Little Saint James surveillance network and the Russian-linked kompromat blackmail videos. - 👤 The Body Double Theory: Examining the Tony Rodham substitution theory and the glaring discrepancies in the official suicide narrative. From the Zorro Ranch secret tunnels to the missing 300GB of Epstein data, we ask the question the mainstream media won't: who is still being protected? This is the ultimate breakdown of systemic failure and the high-stakes world of intelligence assets. 📢 Stop being a spectator to history. Subscribe now and join the search for the missing truth! Give us a 5-star review if you believe in total transparency. 🕵️‍♂️✨  

When studies included in a systematic review have missing outcome data, the study results may be biased and therefore misleading. If there is not much missing data, this is not a problem. If there is lots of missing data, it can be a big problem. Researchers have suggested ways of highlighting how m …