OK. On top of above post.
When you look for SAS examples, don't just look for interrupted time series regression or analysis
, but also for intervention analysis !!
You need a date , time , or date-time variable and that can indeed be just the observation number.
What is your target variable? Is it a continuous outcome, a binary outcome, a count?
Regarding the coding of your intervention : depending on the effect you want to model, you should code the intervention as 0 and 1 (1 as long as the intervention continues) or 0 and 1, 2, 3 (increasing number as long as the intervention continues).
Anyway, you need ARIMAX methodology (dynamic regression) or state space modelling or RNN (Recurrent Neural Networks).
What tooling do you have available?
SAS / ETS ? , SAS Econometrics ? , SAS Forecast Server ? , SAS Visual Forecasting ? , other ? ???
I will answer further after you have answered the above questions.
Koen