TS Covariate Forecast Tool

[rsivasankaran]

I am trying to use the TS Covariate Forecast tool to forecast sales for 31 days in Jan 2023.

The inputs to the TS Covariate Forecast tool are the Output stream from the ARIMA tool and a dataset that has the dates and covariates for each day in Jan 2023. 

I am getting the following error:

   Error: TS Covariate Forecast (53): TS Forecast: Error in `contrasts<-`(`*tmp*`, value = contr.funs[1 + isOF[nn]]) :

I am using 10 years of data until 2022 to train the ARIMA model. 

Is there an example that I could use to understand how to do this?

[rsivasankaran · Accepted solution]

Based on this post it appears that the TS Covariate Forecast tool needs the same input data as the ARIMA tool and it will forecast the same number of forecasts in the input dataset. Since I am interested only in the first 31 days, I will simply ignore the rest of the forecast. 

https://community.alteryx.com/t5/Alteryx-Designer-Discussions/Forecasts-more-than-needed-from-the-TS...