Hello,
I suspect a bug in the forecast when using a gts object. It attached zip file contains the R code in txt format and the data file in csv format. The error thrown by the forecast command is:
Error in x[, (1 + cs[i]):cs[i + 1]] <- xx :
number of items to replace is not a multiple of replacement length
I have been unsuccessful finding assistance with this online and noticed that if I permute the second column of the time series matrix, all works fine but not if the first column is permuted. Thus, I thought there may be an issue with the interaction between the forcast command and the gts object.
Respecfully,
Brad Gubser
bradley.gubser@jmsmucker.com
Hello, I suspect a bug in the forecast when using a gts object. It attached zip file contains the R code in txt format and the data file in csv format. The error thrown by the forecast command is: Error in x[, (1 + cs[i]):cs[i + 1]] <- xx : number of items to replace is not a multiple of replacement length I have been unsuccessful finding assistance with this online and noticed that if I permute the second column of the time series matrix, all works fine but not if the first column is permuted. Thus, I thought there may be an issue with the interaction between the forcast command and the gts object. Respecfully, Brad Gubser bradley.gubser@jmsmucker.com
Issue_in_GTS_forecast.zip