Apply CausalImpact package to the GA output.

causalImpactOnGA.R

## do the causalImpact 
#  here the effect took place on the 200th day of the data imported, so the pre.period = 200
#  Data is examined and compared in the post perid to day 244
impact <- CausalImpact(data = gadata$visits, 
                       pre.period = c(1,200),
                       post.period = c(201,244), 
                       model.args = NULL,
                       bsts.model = NULL, 
                       post.period.response = NULL,
                       alpha = 0.05)

plot(impact)

# this gives you a nice verbose report
impact$report