This site uses different types of cookies, including analytics and functional cookies (its own and from other sites). To change your cookie settings or find out more, click here. If you continue browsing our website, you accept these cookies.
I'm currently trying to build a workflow using a batch macro inside another batch macro, in order to be able to forecast the next year of data of several SKUs and promotions through an ARIMA. The original data base provides a Concatenate of certain info regarding promotions in different stores throughout the year, along with the SKUs promoted at the time. Each promotion includes data from the last week of 2015 to the last week of 2018, although we only want to use the info from 2016-2017, and forecast 2018 (to compare to the real numbers).
What we would like as an outcome is a forecast for the volume of each SKU in each promotion, while keeping clear when it's two different stores/promotions/skus.
Currently the initial workflow assigns a RecordID to each promotion (concatenate) then runs the SKUs batch macros for every RecordID. The SKUs macro gets the list of SKUs of the promotion and uses it on the Arimas batch macro, which runs an ARIMA for each SKU of that promotion. Once all SKUs have their forecast, the next RecordID enters SKUs batch macro and so on.
At the time I've designed both Batch Macros and the outer workflow, but I'm getting error after error when I run it. The most constant error happens when it executes the batch macro that runs the ARIMA, stating the error as "Error in plot.window(...): need finite 'ylim' values".
Does anyone have any idea on how to fix this?
Appreciate any help. I include the workflow + batch macros + data base sample