1. Lalonde's NSW Benchmark | Data Preparation | Deficiency Estimation | 2. Right Heart Catheterization (RHC) | Data Setup | Quantifying the Information Gap | Policy Regret Bounds | Confounding Frontier
Package: causaldef 0.2.0
causaldef: Decision-Theoretic Causal Diagnostics via Le Cam Deficiency
Implements Le Cam deficiency theory for causal inference, as described in Akdemir (2026) <doi:10.5281/zenodo.18367347>. Provides theorem-backed bounds together with computable proxy diagnostics for information loss from confounding, selection bias, and distributional shift. Supports continuous, binary, count, survival, and competing risks outcomes. Key features include propensity-score total-variation deficiency proxies, negative control diagnostics, policy regret bounds, and sensitivity analysis via confounding frontiers.
Authors:
causaldef_0.2.0.tar.gz
causaldef_0.2.0.zip(r-4.7)causaldef_0.2.0.zip(r-4.6)causaldef_0.2.0.zip(r-4.5)
causaldef_0.2.0.tgz(r-4.6-any)causaldef_0.2.0.tgz(r-4.5-any)
causaldef_0.2.0.tar.gz(r-4.7-any)causaldef_0.2.0.tar.gz(r-4.6-any)
causaldef_0.2.0.tgz(r-4.6-emscripten)
manual.pdf |manual.html✨
DESCRIPTION |NEWS
card.svg |card.png
causaldef/json (API)
| # Install 'causaldef' in R: |
| install.packages('causaldef', repos = c('https://denizakdemir.r-universe.dev', 'https://cloud.r-project.org')) |
Bug tracker:https://github.com/denizakdemir/causaldef/issues
Datasets:
- gene_perturbation - Gene Expression Perturbation Data
- hct_outcomes - Hematopoietic Cell Transplantation Outcomes
- nsw_benchmark - Lalonde National Supported Work (NSW) Benchmark
- rhc - Right Heart Catheterization (RHC) Dataset
Last updated from:9973d26f94. Checks:9 OK. Indexed: yes.
| Target | Result | Time | Files | Syslog |
|---|---|---|---|---|
| linux-devel-x86_64 | OK | 209 | ||
| source / vignettes | OK | 225 | ||
| linux-release-x86_64 | OK | 185 | ||
| macos-release-arm64 | OK | 198 | ||
| macos-oldrel-arm64 | OK | 201 | ||
| windows-devel | OK | 278 | ||
| windows-release | OK | 283 | ||
| windows-oldrel | OK | 293 | ||
| wasm-release | OK | 123 |
Exports:audit_datacausal_speccausal_spec_competingcausal_spec_survivalconfounding_frontiercreate_plumber_apicreate_shiny_app_filesestimate_deficiencyestimate_deficiency_competingestimate_effectfrontdoor_effectiv_effectnc_diagnosticoverlap_diagnosticpartial_id_setpolicy_regret_boundpolicy_regret_bound_vcrkhs_rate_boundrun_causaldef_apirun_causaldef_appsharp_lower_boundtest_instrumenttransport_deficiencyvalidate_causal_specwasserstein_deficiency_gaussian
Dependencies:backportscheckmateclicpp11farverggplot2gluegtableisobandlabelinglifecycleR6RColorBrewerrlangS7scalesvctrsviridisLitewithr
Classical Benchmarks: Lalonde and RHC
Last update: 2026-03-23
Started: 2026-02-02
Complete Workflow: From Data to Decision
Overview | Part 1: Gene Perturbation Study (Continuous Outcome) | 1.1 Data Description | 1.2 Step 1: Specification | 1.3 Step 2: Deficiency Estimation | 1.4 Step 3: Diagnose with Negative Control | 1.5 Step 4: Policy Decision | 1.6 Effect Estimation | Part 2: Hematopoietic Cell Transplantation (Survival Outcome) | 2.1 Data Description | 2.2 Step 1: Survival Specification | 2.3 Step 2: Deficiency Estimation | 2.4 Step 3: Confounding Frontier | 2.5 Step 4: Policy Regret and RMST Effect | 2.6 Complete Decision Framework | Part 3: Comparative Analysis Across Studies | 3.1 When Is Observational Evidence Sufficient? | 3.2 General Workflow Summary | References
Last update: 2026-03-23
Started: 2026-02-02
Introduction to causaldef
The Core Workflow | Example: Gene Perturbation Analysis | 1. Specification | 2. Deficiency Estimation | 3. Diagnose with Negative Controls | 4. Decision Making | 5. Effect Estimation
Last update: 2026-03-23
Started: 2026-01-30
Negative Control Diagnostics in causaldef
Introduction | Theoretical Background | What is a Negative Control Outcome? | The Diagnostic Logic | Negative Control Sensitivity Bound (manuscript thm:nc_bound) | Practical Example | Simulating Data with a Negative Control | Creating the Causal Specification | Running the Negative Control Diagnostic | Interpreting the Results | Scenarios | Scenario 1: Adjustment Succeeds | Scenario 2: Adjustment Fails | Choosing Good Negative Control Outcomes | Ideal Properties | Examples by Domain | Combining with Deficiency Estimation | Advanced: Estimating Kappa | Summary | References
Last update: 2026-03-23
Started: 2026-02-02
Sensitivity Analysis: Deficiency vs. E-values
Introduction | Conceptual Translation | The E-value Perspective | The Deficiency Perspective | Conceptual Mapping | Practical Example | Setup | Deficiency Estimation | Confounding Frontier | Policy Regret Bound | Comparison with E-values | Computing an Approximate E-value | Deficiency vs. E-value: Key Differences | When to Use Each | Extended Sensitivity Analysis | Benchmarking Observed Covariates | Combining with Negative Controls | Summary: Unified Sensitivity Analysis | References
Last update: 2026-03-23
Started: 2026-02-02
Survival Analysis with causaldef
Survival Specification | Analysis of HCT Outcomes: Competing Risks | Data Preparation | Specification | Deficiency Estimation | Survival Modeling and RMST | Quantifying Uncertainty: Policy Regret Bounds | Connection to Deficiency
Last update: 2026-03-23
Started: 2026-01-30
The causaldef Methodology: Theory and Practice
Introduction: The Safety Manifesto | Flash Forward: The Destination | The Core Theory: Markov Kernels in Plain English | Theory Overview: The Car and Destination Analogy | Workflow | Step 1: Data Generation | Step 2: Specification | Step 3: Deficiency Estimation | Interpretation of Deficiency Results: The Distance to Walk | Step 4: Diagnostics (The "Negative Control Trap") | Critical Interpretation: The Deductible | Step 5: Effect Estimation | Step 6: Policy Regret Bounds (Transfer Penalty and Safety Floor) | Decision Framework | Step 7: Sensitivity Analysis & Decision | Go/No-Go Decision | Advanced: Survival Analysis | Conclusion
Last update: 2026-03-23
Started: 2026-01-30
Transportability and Policy Learning
1. Transportability: Lalonde's Job Training | Transport Deficiency | 2. Policy Learning Bounds: RHC | Policy Evaluation | The Safety Floor
Last update: 2026-03-23
Started: 2026-02-02
Automated Data Auditing for Causal Studies
Why Audit Your Data? | Case Study: Right Heart Catheterization (RHC) | Understanding the Research Question | Running the Data Audit | Interpreting the Report | Examining Detected Issues | Clinical Interpretation | Comparing Audit Results Across Subsets | Using Audit Results for Causal Analysis | Full Audit Summary | Best Practices for Data Auditing | Conclusion | References
Last update: 2026-02-11
Started: 2026-02-11
Policy Learning with Decision-Theoretic Bounds
Introduction | The Safety Floor Concept | Implications for AI/ML Safety | Practical Workflow | Step 1: Define the Causal Problem | Step 2: Estimate Deficiency | Step 3: Visualize Deficiency | Step 4: Compute Policy Regret Bounds | Step 5: Visualize the Safety Floor | Interpreting the Results | The Safety Floor Report | Sensitivity Analysis with Confounding Frontiers | Visualize the Frontier | Policy Learning with grf (Optional) | Best Practices for Safe Deployment | Pre-Deployment Checklist | Monitoring in Production | Mathematical Details | Policy Regret Transfer (Manuscript) | Why This Matters | Summary | References
Last update: 2026-02-11
Started: 2026-02-02
Advanced Causal Analysis
Introduction | Unobserved Confounding and Negative Controls | Loading the Data | Running the Negative Control Diagnostic | Addressing the Falsification | Sensitivity Analysis: Confounding Frontiers | Visualizing the Frontier | Conclusion
Last update: 2026-01-30
Started: 2026-01-30
