Rolling forecast with xreg and re-estimation

Question

I tried implementing xreg into my rolling forecast with re-estimation. Unfortunately I'm facing issues with the length of xreg.

# sample data  
sample <- ts(rnorm(100, mean = 1000, sd=7), start = c(2012,1), end = c(2019,12), frequency = 12)
external <- ts(mtcars, start = c(2012,1), end = c(2019,12), frequency = 12)

#Define h --> One-step ahead (for a start, later to be increased)
h <- 1
#specify length to forecast
test  <- window(sample, start = c(2018,01), end = c(2019,12), frequency = 12)
n <- length(test) - h + 1
#provide total length of regressors available
total_xreg <- ts(external[,c(1,2,3)], start = c(2012,1), end= c(2019,12), frequency = 12)

#create empty matrix
fcmatx <- matrix(0, nrow=n, ncol=h)

# create loop
for(i in 1:n)
{  
# x is the target variable, provide training data 
  x <- window(sample, end= c(2017,12) + (i-1)/12)
# provide xregs for training data
  xregs <- window(total_xreg, end = c(2017,12) + (i-1)/12)
# provide new xregs for forecasting, assuming that xreg is available for the forecasting period
  xregs2 <- window(total_xreg, start = c(2018,1) + (i-1)/12
# limit xregs2 to show only the first line since we are only forecasting 1 step in advance
 xregs3 <- xregs2[1,]
# create auto.arima model
  refit.multirex <- auto.arima(x, xreg = xregs)
# forecast using regressors
  fcmatx[i,] <- forecast(refit.multirex, 
                         h=h, 
                         xreg = xregs3
                         )$mean
}
fcmattsx <- ts(fcmatx, start = c(2018,1), frequency = 12)

This results in the following error:

Error in forecast.forecast_ARIMA(refit.multirex, h = h, xreg = xregs3) : 
  Number of regressors does not match fitted model

h is length 1 and xregs is length 3 because I'm filling in 3 variables but they are all only for one period of time. I tried various adjustments but cannot get it right.

Answer 1

The following line

xregs3 <- xregs2[1,]

returns a vector rather than a matrix. This is the default behaviour in R when you extract a single column or row from a matrix. Change it to

xregs3 <- xregs2[1,,drop=FALSE]

to keep the matrix structure (1x3). Then the forecast() function will not return an error.

You will get a different error when i=23 because then you have start after end when creating xregs2.