Incorporating Fixed Effects with Causal Impact

Question

I have observations on a dependent variable of interest, $Y_{ijt}$, where $t \in [1,2,...T]$ denotes the time, $i$ denotes a particular brand and $j$ a specific product item. At $t=k$, there was an intervention, corresponding to the opening of a marketing channel used by some of the brands $i$ for some of the product item $j$. I want to use Causal Impact analysis to look at the effect of the marketing channel on the variable $Y$.

If I were to use traditional difference-in-difference analysis to look at the effect, I can incorporate fixed effects corresponding to the brand $i$ and product $j$. How can I incorporate these fixed effects in causal impact analysis?

Answer 1

This is currently not possible with the CausalImpact R package. If it is reasonable to do that in the particular problem, you can aggregate some / all of the items for which a new marketing channel was opened, and check if there is an aggregate effect.