CausalImpact

Dear all, Since its open-source release, the CausalImpact R package has gained fantastic momentum in

Thank you very much! Best, Tina Am Dienstag, 14. April 2015 21:14:10 UTC+1 schrieb Kay Brodersen: Hi

Thanks, I understand is there a way you would write the code for optimization? On Tuesday, April 14,

Hi zedoul, There is no formal upper limit. However, keep in mind that you might run into

Hi yangyang2000, You could construct a custom bsts model and add your metric of advertising intensity

Kay, thank you, I'll try dummy vars! понедельник, 30 марта 2015 г., 11:03:30 UTC+3 пользователь

Actually, you only need to debug(PlotBstsCoefficients), then replace as described. Am Mittwoch, 18.

Hi AJ, One option would be to form the sum of the observed time series of your treated units,

Maybe tried to update RStudio version and R version. My R version is 3.1.2, R Studio version is 0.98.

Hi Israel, Let me check whether I understand your question correctly - do you mean, in a first stage,

Hi Paul, Case 1 and 2 are based on predictors that may have very different flavours from a study-

Dear Ryan, CausalImpact adopts a different approach than the one underlying classical synthetic

You can find a summary of the assumptions in Figure 2 on page 11 of the accompanying manuscript. In

I should have mentioned, also, that the time series are non-stationary with respect to mean and

Dear Kathakali, Yes, that's exactly how I would approach the problem. Note that this may work

Hi Florent, One option would be to align all time series (treatment and control) at the time of

Dear colleagues, Congratulations for this amazing peace of work! If I read well, the package

Hi Piyush, Yes, that's exactly it. Note that for all practical purposes, it's strongly

Thanks Kay! On Friday, January 16, 2015 at 3:25:33 AM UTC-8, Kay Brodersen wrote: Hi Andre, You could

Hi Jitendra, This may be because of an outdated GCC on your machine. Updating GCC to a recent version

Hello, I am trying to calculate sales uplift in a region through this package due to an event. Do you

I am just wondering if anyone in this group has heard of any attempts to look at specific development

As a follow-up, the package documentation now also contains this example. On 8 January 2015 at 12:33,

Hi , Thanks for quick reply . I tried putting that variable in 1/0 format ,but I am getting below

Hi Rodrigo, To debug this, I would check whether you can install other packages from Github. I haven

Hi Antonina, See the previous thread. If the data passed into CausalImpact() is a time-series object

Steve recently uploaded version 0.2 of Boom to CRAN, and that built just fine on my Debian Stable 64-

Hi Jerry, You can plot the contributions of the individual regressors using: plot(impact$model$bsts.

Hi Janelle, I haven't come across this. Could you please provide a minimal reproducible example?

Hi Chenxi, There is no explicit check for collinearity / estimability, but the spike-and-slab prior