Clinical Prediction Suite
ไปป์ไลน์สร้างโมเดลทำนายทางคลินิกด้วย R ครบทั้งกระบวนการตามแนวทาง Riley
A full Riley-style R prediction-model pipeline, end to end.
฿4,990
Not open yet. Leave your email and we’ll tell you the moment it ships.
About this skill
Everything you need to develop and validate a binary clinical prediction model in R, the way the Riley/TRIPOD playbook prescribes. It ships final logistic-regression and random-forest models, internal-validation and model-instability bootstraps, and a matched SHAP explainability pair. Data harmonization, matching methods, and a reproducible graph gallery round out the same toolset used in real prediction-model studies.
What’s inside
- Final logistic-regression prediction model (glm + elastic net)
- Final random-forest model (tidymodels + ranger)
- Internal-validation bootstrap (optimism-corrected performance)
- Model-instability bootstrap with stability plots
- SHAP explainability for both logistic and random-forest models
- Data harmonization and variable-degree checks
- Greedy and prognosis-score matching helpers
- Reproducible ggplot2 graph gallery