Prediction binary outcome, Measuring prediction error: AUC / Brier Score

collapsibility example in epidemiological setting

Advantage of backward elimination over forward selection in regression methods

Cross validation vs bootstrap

interaction vs effect measure modification in observational epidemiological studies (w/confounding)

Reference for machine learning topics

Critical Appraisal of Published Articles using Machine Learning methods in clinical research

Model Development Considerations for Machine Learning Implementations in Clinical Applications

Unsupervised Learning: K-means, Optimal number of clusters, choose different initial means (nstart)

Overview of supervised learning in machine learning

Prediction binary outcome, Measuring prediction error: AUC / Brier Score

Prediction continuous outcome, Design Matrix, Measure prediction error: rmse/ r2, Overfit/ Optimism

Supervised Learning: CART vs Ensemble method (bagging/ boosting/ Super Learner); Variable importance

RHC data description

Cross validation in machine learning: k-fold, 1-by-1, automated caret package, performance measure

Data splitting in machine learning; test, train and performance measures

R Guide for TMLE in Medical Research (R/Medicine Conference Short Course)

Lab 4h Supopulation in Complex Survey data (Subsetting)

Lab 4g Analyzing another the analytic dataset from NHANES

Lab 3d definition of collapsibility, and examples in RD, RR and OR: marginal & conditional estimates

Lab3c problem with change-in-estimate method for odds ratios

presenting at a conference or a seminar

RStudio + @GitHub: Manuscript Writing Collaboration Tools (Bookdown site + PDF via Rmarkdown)

Scientific writing Discussion Section

Scientific writing Results Section