Date:         Tue, 27 Mar 2012 11:31:40 -0300
Reply-To:     Ricardo Gonçalves da Silva
              <ricardo.gsilva@BANCOVOTORANTIM.COM.BR>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Ricardo Gonçalves da Silva
              <ricardo.gsilva@BANCOVOTORANTIM.COM.BR>
Subject:      RES: Time series
In-Reply-To:  <201203271345.q2RA75F6028469@waikiki.cc.uga.edu>
Content-Type: text/plain; charset="iso-8859-1"

Hi William,

I think that the best way is to test the original series for stationarity (Dickey-Fuller related testes) and then use the traditional modeling techniques.
However, if you have intervention, I suppose you need two things to do:

1) Traditional intervention analyses: http://support.sas.com/documentation/cdl/en/etsug/60372/HTML/default/viewer.htm#etsug_arima_sect013.htm
2) Testing for unit root with breaking trends and variance.

Rick


-----Mensagem original-----
De: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU] Em nome de William Shakespeare
Enviada em: terça-feira, 27 de março de 2012 10:45
Para: SAS-L@LISTSERV.UGA.EDU
Assunto: Time series

I have some time series data that consist of measurements on individuals at one time point (length of treatment, total costs in constant dollars).
The data set has several years of such measurements and my question concerns pre- and post-intervention trends in the outcomes (with all the caveats about rival hypotheses).

I'm not sure whether to treat the individuals as the unit of analysis or to average everyone's outcomes on a monthly basis.  Not having done a lot of time series analysis I'm not sure what questions I need to answer in order to come to a conclsion about whether or not to average.

I know in economic data concepts like stationarity are important and analyses are typically of avaerage or total sales, so that seems to imply some monthly average and differencing is required.

Can someone help me out?

Esta mensagem e seus anexos podem conter informações confidenciais ou privilegiadas. Se você não é o destinatário, não está autorizado a utilizar o material para qualquer fim. Solicitamos que você apague a mensagem e avise imediatamente o remetente. O conteúdo desta mensagem e seus anexos não representam necessariamente a opinião e a intenção da empresa, não implicando em qualquer obrigação ou responsabilidade por parte desta.

This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. The contents of this message and its attachments do not necessarily express the opinion or the intention of the company, and do not imply any legal obligation or responsibilities from this company.