Forum Discussion
Using correlation to predict/estimate missing values
HI Jay_Lu,
I suppose the ranking formulas are calculated based on the detail level records, when you use it on table visual these records have been aggregated the sales values with the current category.
For this scenario, I think you may need to change the formula to ranking records based on group. You can take a look at the following blog if helps:
RANKX on multiple columns with DAX and Power BI - SQLBI
Regards,
Xiaoxin Sheng
Hi Xiaoxin,
Thanks for your suggestion, but my case would be different.
So the ranking is actually from an extenral source, and I lookupvalue the sales via the person's name.
By saying that, we know all people's ranking, and some salespersons' sales firgues, and now we want to estimate those missing sales with confirmed rankings.
I initially thought to calculate the correlation between known rankings and turnover, and then use a model to predict the missing values. But just don't know how to. maybe there is other way