Multivariate Autoregressive State-Space Modeling
Return accuracy metrics
MARSS Function Defaults and Allowed Methods
Convert Model Objects between Forms
Check inputs to MARSS call
Check model List Passed into MARSS Call
Coefficient function for MARSS MLE objects
Plot Extinction Risk Metrics
Plot Forecast Uncertainty
Example Data Sets
Describe a marssMODEL Objects
Return fitted values for X(t) and Y(t) in a MARSS model
forecast function for marssMLE objects
Return brief summary information on a MARSS fit
Tests marssMLE object for completeness
Test Model Objects
Return a diagonal list matrix
logLik method for MARSS MLE objects
Multivariate Autoregressive State-Space Model Estimation
Fit a MARSS Model via Maximum-Likelihood Estimation
Multivariate Dynamic Factor Analysis
Multivariate AR-1 State-space Model
Multivariate AR-1 State-space Model with Inputs
Vectorized Multivariate AR-1 State-space Model
AIC for MARSS Models
Names for marssMLE Object Components
Bootstrap MARSS Parameter Estimates
MARSScv is a wrapper for MARSS that re-fits the model with cross valid...
Observed Fisher Information Matrix at the MLE
Generic for fitting MARSS models
Hessian Matrix via the Harvey (1989) Recursion
Compute Expected Value of Y, YY, and YX
Parameter Variance-Covariance Matrix from the Hessian Matrix
Hessian Matrix via Numerical Approximation
MARSS Error Messages and Warnings
Initial Values for MLE
Bootstrapped Data using Stoffer and Wall's Algorithm
EM Algorithm function for MARSS models
Model Checking for MLE objects Passed to MARSSkem
Kalman Filtering and Smoothing
Class "marssMLE"
Class "marssMODEL"
Parameter estimation for MARSS models using optim
Standard Errors, Confidence Intervals and Bias for MARSS Parameters
Class "marssPredict"
Class "marssResiduals"
MARSS Residuals
MARSS Smoothed Residuals
MARSS One-Step-Ahead Residuals
MARSS Contemporaneous Residuals
Simulate Data from a MARSS Model
Vectorize or Replace the par List
match.arg with exact matching
model.frame method for marssMLE and marssMODEL objects
Plankton Data Sets
Plot MARSS MLE objects
Plot MARSS Forecast and Predict objects
Plot MARSS marssResiduals objects
predict and forecast MARSS MLE objects
predict and forecast MARSS MLE objects
Printing functions for MARSS MLE objects
Printing marssMODEL Objects
Printing function for MARSS Predict objects
Model and state fitted values, residuals, and residual sigma
Standardized Innovations
Summary methods for marssMLE objects
Palettes
Return estimated parameters with summary information
Create a LaTeX Version of the Model
Smoothed and filtered x and y time series
Utility Functions
z-score a vector or matrix
The MARSS package provides maximum-likelihood parameter estimation for constrained and unconstrained linear multivariate autoregressive state-space (MARSS) models, including partially deterministic models. MARSS models are a class of dynamic linear model (DLM) and vector autoregressive model (VAR) model. Fitting available via Expectation-Maximization (EM), BFGS (using optim), and 'TMB' (using the 'marssTMB' companion package). Functions are provided for parametric and innovations bootstrapping, Kalman filtering and smoothing, model selection criteria including bootstrap AICb, confidences intervals via the Hessian approximation or bootstrapping, and all conditional residual types. See the user guide for examples of dynamic factor analysis, dynamic linear models, outlier and shock detection, and multivariate AR-p models. Online workshops (lectures, eBook, and computer labs) at <https://atsa-es.github.io/>.