each observation is a patient?
xtset i(patientid) t(month)
xtreg antibiotic group
adjust, by(group) se ci
then repeat the analysis if time<inttervention month, and for
time>intervention month, to see if group differences developed after the
intervention
-------------------------------------------------------------------
George M Hoffman MD
Children's Hospital of Wisconsin
Associate Professor, Anesthesiology and Pediatrics
Medical College of Wisconsin
(414)-266-3560
-------------------------------------------------------------------
----- Original Message -----
From: "Thursky, Karin A" <[email protected]>
To: <[email protected]>
Sent: Monday, March 28, 2005 7:33 AM
Subject: st: short time-series analysis
I have a data on antibiotic usage per month (as grams) for 6 months before
and 6 months after an intervention. The Cochrane group recommend repeated
measures analysis, multiple t-tests and analysis of variance as accepted
methodology for short time-series. (I assume that I have insuffucient data
for an interrupted time series regression using ARIMA)
The data includes
- group (intervention/non-intervention)
- antibiotic in grams
- month
Can anyone suggest a stata methodology that is suitable for a short
time-series analysis?
Regards
Karin
Karin Thursky
Infectious Diseases Physician
Royal Melbourne Hospital/Peter MacCallum Cancer Institute
[email protected]
0414 996649
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
