When Y has class "mts"

[jacobbien]

The line

if(class(Y)!="matrix"){Y=matrix(Y,ncol=1)}

assumes that class(Y) has length 1. However, commonly we have

class(Y)
[1] "mts" "ts" "matrix"

and applying "as.matrix" as suggested in https://arxiv.org/pdf/1702.07094.pdf on page 10 doesn't convert this to have class "matrix" since this is an inherited class from "matrix"...

class(as.matrix(Y))
[1] "mts" "ts" "matrix"

So it leads to a warning when the above "if" statement is called, and worse -- it converts the multivariate ts into a (nrow(Y) * ncol(Y))-by-1 matrix.

[wbnicholson]

I changed the class(Y)!="matrix" calls to is.vector(Y). That should resolve this issue, let me know if it doesn't.

[Neil-Schneider]

Doesn't it make more sense to look for matrix in the class list?

if (!"matrix" %in% class(Y)) {
  ...
}

A time series object is not a vector and assume you would want to convert those to a matrix also.

tmp = ts(c(1,2,3,4,5))
is.vector(tmp)

[jacobbien]

Yes, exactly. I agree that

!"matrix" %in% class(Y)

is better.

[wbnicholson]

I've updated all is.vector tests to !"matrix" %in% class(Y). These errors should be resolved.

[bdemeshev]

For me the following code still fails (I've tested both CRAN and github versions):

library(BigVAR)

data(Y)
y_st <- apply(Y, 2, FUN = scale)
y_st_ts <- ts(y_st, start = 1, frequency = 1)

model_spec <- constructModel(y_st_ts, p = 1, 
                             struct = "OwnOther", 
                             ownlambdas = TRUE,
                             gran = 10)
model_est <- BigVAR.est(model_spec)

Without ts() function everything seems fine!