Stata package for Interrupted Time Series Impact Evaluation Design

I heard this design in my 25 unit course on impact evaluation and immediately got interested with it. The idea is simple.  Basically, in the absence of an intervention (which is the interruption in the time series data set), the outcome variable would just follow the trend.  Sort of like, history repeating itself.  However, with an intervention, there would be marked interruption in the trend.

How do you implement interrupted time series?  I recently found that Stata has a user developed command packaged called itsa (Interrupted Time Series Analysis).  

According to the package description, "itsa estimates the effect of an intervention when the outcome variable is ordered as a time series, and a number of observations are available in both pre- and post-intervention periods. The study design is generally referred to as an interrupted time series because the intervention is expected to interrupt the level and/or trend subsequent to its introduction (Campbell & Stanley, 1966; Glass, Willson & Gottman, 1975; Shadish, Cook & Campbell, 2002)"

Let us try this lovely Stata command.  Type the following in your command panel:

1. sysuse cigsales_single, clear // this opens the data set
2. tsset year // in stata, it is important to declare the dataset as time series
3. itsa cigsale, single trperiod(1989) lag(1) fig posttrend //this implements the itsa, the single trperiod (1989) is the introduction of the intervention, while fig creates the graph while posttrend computes for post-intervention linear trend.