I have a statistics question regarding a regression model with the following log transformed continuous variables.
DV = spend -- dollars spent
IVs = purchases -- number of purchases
aov -- average order value
The tricky part is that aov is a calculated variable = spend / purchases. I have been reading about interaction effects with continuous variables but not using the interaction in this way. The aov is a function of both spend and purchases but it represents a unique value, not spend*purchases. Any light shed on this topic is much appreciated.
Locked
