Chapter 9: Model-Based Feature Extraction

Bayesian Inference and MCMC for Time Series advanced

A Bayesian time-series model produces a posterior distribution, not just a fitted line, which is why posterior uncertainty can itself become a feature.

Register to Read

Create Free Account

Already have an account? Sign in

References

The No-U-Turn Sampler: Adaptively Setting Path Lengths in Hamiltonian Monte Carlo

Matthew D. Hoffman, Andrew Gelman (2011) — arXiv:1111.4246 [cs, stat]

Understanding the Metropolis-Hastings Algorithm

Siddhartha Chib, Edward Greenberg — The American Statistician

Markov Chain Monte Carlo and Variational Inference: Bridging the Gap

Tim Salimans, Diederik P. Kingma, Max Welling (2015) — arXiv:1410.6460 [stat]

Chapter

9 Model-Based Feature Extraction

More Topics

Autoregressive, Moving-Average, and ARIMA Foundations for Feature Engineering Fractional Differencing: Keeping Memory Without Keeping the Unit Root Path Signatures and Log-Signatures for Financial Sequences Regime Models for Feature Engineering: HMMs, Markov Switching, and Distributional Clustering State-Space Models and Kalman Filtering for Feature Engineering Stationarity Tests: ADF, KPSS, and Rolling Stability Signals Structural Break Diagnostics and Time-Since-Break Features Uncertainty as a Feature Volatility Models as Feature Extractors: GARCH, EGARCH, and HAR Wavelets for Multi-Scale Diagnostics and Causal Feature Design