sigex.cast: Computes forecasts and aftcasts, without uncertainty; a...

In jlivsey/sigex: Ecce Signum: Multivariate Signal Extraction

Computes forecasts and aftcasts, without uncertainty; a faster version of sigex.midcast, if you don't need midcasts

Description

Background: psi refers to a vector of real numbers containing all hyper-parameters (i.e., reals mapped bijectively to the parameter manifold)

Usage

sigex.cast(psi, mdl, data.ts, leads)

Arguments

psi	A vector of all the real hyper-parameters
mdl	The specified sigex model, a list object
data.ts	A T x N matrix ts object (with no missing values) corresponding to N time series of length T
leads	An integer sequence of desired casts; include index t to obtain an estimate of x_t. These integers don't have to be a subset of 1,2,...,T. Include integers greater than T to get forecasts, or less than 1 to get aftcasts.

Details

Notes: presumes that regression effects have already been removed.

Value

x.casted: N x (T+H) matrix of aftcasts, data, and forecasts, where H is the total number of forecasts and aftcasts

jlivsey/sigex documentation built on March 20, 2024, 3:17 a.m.