What statistical languages does this skill support?

This skill primarily supports Stan and JAGS, which are the industry standards for Bayesian statistical modeling and Markov Chain Monte Carlo (MCMC) sampling.

Does it include state-space models?

Yes, it provides implementation patterns for Local Level (Random Walk + Noise) and Local Linear Trend models, which are foundational for Bayesian state-space analysis.

How are seasonal effects handled in these models?

The skill provides patterns for additive seasonality using sum-to-zero constraints, allowing you to model periodic fluctuations in time series data effectively.

Can this skill help with financial forecasting?

Yes, it includes specific models like GARCH(1,1) and AR(p) which are widely used for modeling volatility, asset returns, and financial trends.

Bayesian Time Series Models

Name: Bayesian Time Series Models
Author: choxos

bychoxos

데이터 과학 및 ML

Implements advanced Bayesian time series analysis using Stan and JAGS for probabilistic forecasting and state-space modeling.

소개

The Bayesian Time Series Models skill equips Claude with specialized knowledge and implementation patterns for complex temporal data analysis. It provides production-ready code for Autoregressive (AR) models, state-space models like Local Linear Trend, and volatility clustering with GARCH. By leveraging Stan and JAGS, this skill allows users to perform rigorous probabilistic inference, handle missing data, and incorporate prior domain knowledge into time-series forecasting workflows, ensuring robust statistical conclusions even with limited or noisy data.

주요 기능

Seasonal decomposition using sum-to-zero additive seasonal effects
Optimized vectorized Stan code for efficient Hamiltonian Monte Carlo sampling
0 GitHub stars
GARCH(1,1) patterns for modeling volatility clustering in financial data
State-space model implementations including Local Level and Linear Trend
Templates for AR(1) and AR(p) models with stationarity constraints

사용 사례

Predicting financial market volatility and asset returns using GARCH models
Filtering noisy sensor data using Bayesian random walk and state-space frameworks
Analyzing seasonal trends in business metrics and consumer demand forecasting

소개

주요 기능

Seasonal decomposition using sum-to-zero additive seasonal effects
Optimized vectorized Stan code for efficient Hamiltonian Monte Carlo sampling
0 GitHub stars
GARCH(1,1) patterns for modeling volatility clustering in financial data
State-space model implementations including Local Level and Linear Trend
Templates for AR(1) and AR(p) models with stationarity constraints

사용 사례

Predicting financial market volatility and asset returns using GARCH models
Filtering noisy sensor data using Bayesian random walk and state-space frameworks
Analyzing seasonal trends in business metrics and consumer demand forecasting