MARSS-package: Multivariate Autoregressive State-Space Model Estimation

Description Details Author(s) References

Description

The MARSS package fits constrained and unconstrained multivariate autoregressive time-series models to multivariate time series data. To open the user guide from the command line, type RShowDoc("UserGuide",package="MARSS"). To open an overview page with package information and links to the R scripts in the User Guide, type RShowDoc("index",package="MARSS").

The main function is MARSS which is used to fit a specified model to data and estimate the model parameters. MARSS model specification is based on "form" (an argument to a MARSS() call). The form tells MARSS() what to expect in the model list (model is a MARSS argument) and how to translate that into the base model form used in the fitting algorithms.

The default MARSS model form is "marxss", which is a model of the following form:

x(t+1) = B x(t) + U + C c(t) + w(t), where w(t) ~ MVN(0,Q)
y(t) = Z x(t) + A + D d(t) + v(t), where v(t) ~ MVN(0,R)
x(1) ~ MVN(x0, V0) or x(0) ~ MVN(x0, V0)

The parameters, hidden state processes (x), and observations (y) are matrices:

All parameters can be time-varying.

The package functions estimate the model parameters using an EM algorithm (primarily but see MARSSoptim). Parameters may be constrained to have shared elements (elements which are constrained to have the same value) or fixed elements (with the other elements estimated). The states and smoothed state estimates are provided via a Kalman filter and smoother. Bootstrapping, confidence interval estimation, bias estimation, model selection and simulation functions are provided. The main user interface to the package is the top-level function MARSS.

Details

Important MARSS functions:

MARSS

Top-level function for specifying and fitting MARSS models.

MARSSsimulate

Produces simulated data from a MARSS model.

MARSSkem

Estimates MARSS parameters using an EM algorithm.

MARSSkf

Kalman filter and smoother.

MARSSoptim

Estimates MARSS parameters using a quasi-Newton algorithm via optim.

MARSSaic

Calculates AICc, AICc, and various bootstrap AICs.

MARSSboot

Creates bootstrap MARSS parameter estimates.

MARSSparamCIs

Computes confidence intervals for maximum-likelihood estimates of MARSS parameters.

Author(s)

Eli Holmes, Eric Ward and Kellie Wills, NOAA, Seattle, USA.

eli(dot)holmes(at)noaa(dot)gov, eric(dot)ward(at)noaa(dot)gov,

kellie(dot)wills(at)noaa(dot)gov

References

The MARSS user guide: Holmes, E. E., E. J. Ward, and M. D. Scheuerell (2012) Analysis of multivariate time-series using the MARSS package. NOAA Fisheries, Northwest Fisheries Science Center, 2725 Montlake Blvd E., Seattle, WA 98112 Type RShowDoc("UserGuide",package="MARSS") to open a copy.

Type RShowDoc("index",package="MARSS") to see all the package documentation, tutorials, and R scripts from the User Guide.


gragusa/MARSS documentation built on May 17, 2019, 8:18 a.m.