GroupVARX: Group Vector Autoregression with Exogenous Inputs via Group...

In jeffwong/fastVAR: fastVAR

Description

Fit a VAR model by creating the lagged design matrix and fitting a multivariate response matrix to it. Note that the GroupVARX response matrix omits the first max(p,b) responses from the input matrix Y. This is because observations in Y are modeled by the max(p,b) previous values, so the first max(p,b) observations cannot be modeled. While multivariate response regressions can be solved as multiple univariate response regressions, this multivariate response problem can better be solved by using Group Lasso. Instead of seeking sparsity in the coefficients for each univariate response, Group Lasso attempts to find sparsity in the coefficient matrix as a whole.

Usage

  GroupVARX(y, freq = rep(NA, ncol(y)), x, p = 1, b = 1,
    weights = NULL, alpha = 0.4, getdiag = T)

Arguments

y	A matrix of endogenous variables where each column represents an individual time series
freq	only used if the time series are periodic. freq is a vector of frequencies for each of the time series, as in 'ts(y, freq = ...)'. If the time series are not periodic, then this vector can be a vector of NA
x	A matrix of exogenous variables where each column represents an individual time series
p	the number of lags of Y to include in the design matrix
b	the number of lags to X include in the design matrix
weights	weights applied to the multiresponse linear regression. Better predictions might come from weighting observations far in the past less so they impact the objective value less. Either NULL, "exponential", or "linear"
alpha	the mixing parameter between group lasso and quadratic, as in 'glmnet'
getdiag	logical. If true, return diagnostics

Examples

data(Canada)
  x = matrix(rnorm(84*4), 84, 4)
  GroupVARX(y, x=x, p=3, b=2)

jeffwong/fastVAR documentation built on May 19, 2019, 4:02 a.m.