2 Replies Latest reply on Nov 18, 2016 6:41 PM by Bora Beran

    Bi-variate analysis without R

    Frank Casella

      Hi,

       

      Is there a way to perform a multivariate analysis in Tableau without using an external software suite?  I'm trying to do a bivariate forecast where customer segment and time are the independent variables and sales is the dependent variable.  Basically, our customer mix is changing and I want to forecast how this could affect future sales. I've got 30 months of sales data broken down by segment to work with.

       

      I can brute force my way through it by creating a forecast for each of the segments independently, exporting then joining the data, and then performing a weighted aggregation of the forecast. I'm doing this by creating separate regression equations for each segment since I can't figure out how to perform a customer segment level of detail calculation since the regression is a table calculation.

       

      I figure there has to be an easier way. Any suggestions would be greatly appreciated!

       

      Thanks,

       

      Frank