PyMC Bayesian Modeling
技能 已验证 活跃Bayesian modeling with PyMC. Build hierarchical models, MCMC (NUTS), variational inference, LOO/WAIC comparison, posterior checks, for probabilistic programming and inference.
To enable AI agents to perform advanced Bayesian statistical modeling and probabilistic programming tasks using PyMC.
功能
- Build and fit Bayesian models with PyMC
- Implement hierarchical models and non-centered parameterization
- Perform MCMC sampling (NUTS) and variational inference (ADVI)
- Conduct prior and posterior predictive checks
- Compare models using LOO and WAIC
- Analyze sampling diagnostics and troubleshoot issues
使用场景
- Building complex hierarchical models for grouped data
- Performing uncertainty quantification through Bayesian inference
- Validating model assumptions and fit using predictive checks
- Comparing multiple statistical models to find the best fit
- Implementing advanced statistical analyses in research workflows
非目标
- Performing basic frequentist statistical tests
- Building machine learning models not based on Bayesian principles
- Automating data collection or external API calls beyond model parameterization
工作流
- Prepare and standardize data
- Build the Bayesian model structure with PyMC
- Perform prior predictive checks
- Fit the model using MCMC or VI
- Check sampling diagnostics (R-hat, ESS, divergences)
- Validate model fit with posterior predictive checks
- Analyze model parameters and make predictions
实践
- Bayesian modeling workflow
- Prior selection and validation
- Model diagnostics and convergence checking
- Hierarchical model construction
- Model comparison and selection
先决条件
- Python 3.11+
- PyMC installed
- ArviZ installed
安装
npx skills add K-Dense-AI/claude-scientific-skills通过 npx 运行 Vercel skills CLI(skills.sh)— 需要本地安装 Node.js,以及至少一个兼容 skills 的智能体(Claude Code、Cursor、Codex 等)。前提是仓库遵循 agentskills.io 格式。
质量评分
已验证类似扩展
Alterlab Pymc
98Part of the AlterLab Academic Skills suite. Bayesian modeling with PyMC. Build hierarchical models, MCMC (NUTS), variational inference, LOO/WAIC comparison, posterior checks, for probabilistic programming and inference.
Simulate Stochastic Process
97Simulate stochastic processes (Markov chains, random walks, SDEs, MCMC) with convergence diagnostics, variance reduction, and visualization. Use when generating sample paths for estimation, prediction, or visualization; when analytical solutions are intractable; running Monte Carlo estimation needing convergence guarantees; validating analytical results against empirical simulation; or sampling from complex posteriors via MCMC.
Fit Drift Diffusion Model
100Fit cognitive drift-diffusion models (Ratcliff DDM) to reaction time and accuracy data with parameter estimation (drift rate, boundary separation, non-decision time), model comparison, and parameter recovery validation. Use when modeling binary decision-making with reaction time data, estimating cognitive parameters from experimental data, comparing sequential sampling model variants, or decomposing speed-accuracy tradeoff effects into latent cognitive components.
LLM Models via OpenRouter
99Access Claude, Gemini, Kimi, GLM and 100+ LLMs via inference.sh CLI using OpenRouter. Models: Claude Opus 4.5, Claude Sonnet 4.5, Claude Haiku 4.5, Gemini 3 Pro, Kimi K2, GLM-4.6, Intellect 3. One API for all models with automatic fallback and cost optimization. Use for: AI assistants, code generation, reasoning, agents, chat, content generation. Triggers: claude api, openrouter, llm api, claude sonnet, claude opus, gemini api, kimi, language model, gpt alternative, anthropic api, ai model api, llm access, chat api, claude alternative, openai alternative
Generate Statistical Tables
99Generate publication-ready statistical tables using gt, kableExtra, or flextable. Covers descriptive statistics, regression results, ANOVA tables, correlation matrices, and APA formatting. Use when creating descriptive statistics tables, formatting regression or ANOVA output, building correlation matrices, producing APA-style tables for academic papers, or generating tables for Quarto and R Markdown documents.
Strategy Validator
99使用 4 种统计测试（DSR、Walk-Forward、Regime、CPCV）来验证交易策略是否存在过度拟合