I have a Dataset in Tableau, from which i need to get the Slope and Intercept of the Linear Regression best fit line using the lm() function in R.
The regression is a basic one, with just one predictor variable and one dependent. How do I handle the fact that i need to pass an aggregated value to the lm() function, even though i need to send the entire column?
With a column called 'Predictor', and another called 'Dependent', I essentially need to do this in Tableau:
fit <- lm(df$Dependent ~ df$Predictor)
return fit
Tableau does not let me do this unless I pass to it
SUM[Dependent] and SUM[Predictor]
But instead of aggregating it, i need it to work on the entire column
Thanks
Add a dimension to any shelf, say the detail shelf, that partitions the data at the level you wish. If you want each individual row to be included in the model, use a dimension that has a unique value for each row, say some sort of row id.
BTW, you know you can perform regression models directly within Tableau, right? You don't need to call out to R just to fit a linear model.