Hi Carolina,
The "transfer function" is really old school. Stata allows you to enter
covariates in ARIMA, which it refers to as "ARMAX" which allows you the
flexibility to model treatment effects in time series models.
Therefore you can generate a covariate for the "step" (the value of the
outcome in the month right after the start of the intervention), as well as
the "post intervention trend" (and interaction term between the "time"
variable and "treatment" variable).
Ariel
Date: Sun, 3 Mar 2013 16:39:54 -0300
From: Caro ALIYA <aliyabay@gmail.com>
Subject: st: Interrupted Time Series Analysis
Dear all,
Im running an ARIMA model with an intervention that is gradual and
permanent.
Is there a way in Stata to specified the transfer functions (parameter
w & delta).
Thanks,
Carolina
*
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/faqs/resources/statalist-faq/
* http://www.ats.ucla.edu/stat/stata/
