Skip to Content

Error: Date variable must be strictly increasing

Sep 26, 2017 at 04:33 PM


avatar image

Hi experts:

Getting error msg during the time series analysis as there are two values associated with same date. I am taking data from HANA how can we aggregate the dates ?? should that be handled on modelling side?? or can we do on predictive Modeler side of Predictive tool to get rid of the error

10 |10000 characters needed characters left characters exceeded
* Please Login or Register to Answer, Follow or Comment.

1 Answer

Best Answer
Sep 28, 2017 at 02:07 PM

Hi Neeraja,

does it make sense (from a business standpoint) to aggregate the two values corresponding to the date?

Kind regards,


Show 2 Share
10 |10000 characters needed characters left characters exceeded

Hi all,

I have the same issue.

if we have many product and customer invoiced each day, so we can't aggregate by date, it doesn't make sense.

If we want to have CustomerId, ProductID as variables in a Time Series analysis, how can perform within Predictive Analysis ?



Hi Pierrick, can you please describe in detail what you want to achieve? If you want to forecast a quantity that depends on both customer and product variables you need to organize your dataset as follows:

Customer ID Product ID Date Quantity

Then specify the multiple rows.

The tool is saying that per segment (here each segment is a combination or crossing of customer * product) you need to arrange the rows so that the date is strictly increasing which ultimately is a matter of data sorting per segment not aggregation.

Kind regards Antoine