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've just done some testing and there are two key things to fix here.
Firstly, the note at the bottom 'The input data must have data for at least a year plus the number of periods prior to this date, for example, for weekly data with 2 periods selected, you need 54 prior records in your data stream.'
This criteria is met in very few (I think only one instance, the last one) in your workflow.
Secondly, your data is at a level of aggregation beyond that needed for the tool. You want to predict the gross-margin by store. But actually your dataset is aggregated at day-invoice-week-store.
You can drop a summerize tool in before your AB-Trend tool, and then summerize by Store, Week and then aggregate the margin value in the way you want, say sum.
Then you can plug this in, and take the 'year of prior records' note into account and you should be good!
I am working on a Time Series problem wherein I am trying to impute the missing values in a dataset. I have been thinking of using the Trend and Seasonality present in the data for this imputation exercise which I am trying to calculate using the AB Trend Tool.
Can you comment on this approach or if I should be adopting some other approach to impute the missing values in the dataset.