User guide

The user guide explains how to do econometrics with cultivars, one family of models or one task at a time. Each page gives the model in its standard notation, runs a verified example, lists the specification choices that matter, and ends with the nuances that are easy to get wrong. The API reference has the exhaustive argument lists; the guide is where to learn which class to reach for and how its parts fit together.

New to the library? Read Quickstart first, then Architecture and Conventions. Those two pages explain the object model and the shape and index rules that every other page assumes.

Find the right page

Question	Page
How are models, results, schemes and evaluators related?	Architecture
What shape do inputs take? How are returned arrays indexed?	Conventions
Which model fits my data and question?	Choosing a model
What behaves differently from what I would expect?	Nuances and known issues
Is my series stationary, cointegrated, nonlinear, broken?	Diagnostics and specification testing
AR, ARIMA, seasonal ARIMA, ARFIMA, unobserved components	Univariate linear models
GARCH, GJR, EGARCH, FIGARCH, stochastic volatility	Volatility models
Threshold, smooth-transition, Markov-switching, neural AR	Nonlinear and regime-switching univariate models
VAR and VARX: estimation, inference, IRFs, forecasts	Vector autoregressions
VECM, VARMA, panel, global, mixed-frequency, functional, yield curve	Beyond the basic VAR
Recursive, long-run, sign, narrative, proxy and other identification	Structural identification
Bayesian VARs and their priors	Bayesian VARs
Factor models, FAVAR, sparse and network VARs, spillovers	Large and high-dimensional systems
Threshold, smooth-transition, time-varying and switching VARs	Nonlinear, time-varying and regime-switching VARs
Kalman, particle and regime-switching filters	State-space models and filtering
Spectra, band-pass and trend filters, wavelets	Spectral analysis and filtering
Backtests, scoring rules, forecast comparison tests	Forecasting and forecast evaluation
Convergence, predictive checks, marginal likelihood, model averaging	The Bayesian workflow

Foundations

• Architecture
• Conventions
• Choosing a model
• Nuances and known issues

Before estimation

• Diagnostics and specification testing

Univariate models

• Univariate linear models
• Volatility models
• Nonlinear and regime-switching univariate models

Multivariate models

• Vector autoregressions
• Beyond the basic VAR
• Structural identification
• Bayesian VARs
• Large and high-dimensional systems
• Nonlinear, time-varying and regime-switching VARs

Engines and frequency-domain views

• State-space models and filtering
• Spectral analysis and filtering

After estimation

• Forecasting and forecast evaluation
• The Bayesian workflow