Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: Dealing with an unusual peak in time-series data
From
Ronan Conroy <[email protected]>
To
"[email protected]" <[email protected]>
Subject
Re: st: Dealing with an unusual peak in time-series data
Date
Thu, 3 Feb 2011 17:49:46 +0000
On 3 Feb 2011, at 17:41, Jung-eun Lee wrote:
> Hi all,
> I have data on monthly counts of certain events for about 20-year period.
> I looked at the temporal pattern of the monthly counts, and I found a HUGE peak in the middle of the data (for 2-3 months).
> I wonder whether or not it is ok to estimate a time-series model like negative binomial regression (this data has lots of zeros) when the data looks like this. Any other model suggestion?
This suggests that a process was at work for this time period which did not apply to the rest of the series. You are better off trying to identify what this was (human error?) and deal with it as a separate issue. It may not be appropriate to model this period with the remainder of the data.
Ronán Conroy
[email protected]
Associate Professor
Division of Population Health Sciences
Royal College of Surgeons in Ireland
Beaux Lane House
Dublin 2
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/