--- In [email protected], Peter <[EMAIL PROTECTED]> wrote:
>
> Akasha, did they use an interupted time series
> analysis? I'm assuming with my baby stats background
> that this would have been appropriate.  

Yes, it appears they used an ARIMA / Box-Jenkins model. Which is
efficient for impact anlysis with highly seasonal and autocorrelated
data -- typically financial data. The crime data may have seasonality
but does not appear to be that strongly autocorrelated. And
traditioanlly, it requires 5-6 seasons of data to be reliable. Since
one of their models was only for 1993 data (from the summary) it
raises some questions. 

IMO, its too bad they did not use a more generalized regression
approach. Anything that can be done in ARIMA can be done in a
regression model specification -- by using differening and lagging of
variables, and use of dummy variables for impacts and seasonality.
ARIMA, IMO, is a bit of a black box, regression models are much more
transparent. And regression allows, at least is easier for, testing a
large number of independent control variables in a unified model. 

The summary states that temperature was the only control variable used
in the primary analysis for the 1 year and 5 year analysis. Then,
later, they tested a range of social / LE control variables to show,
per their satisfaction, that the contol variables were insignificant
in effecting the results of the 1 and 5 year models. Aside from being
methodologically weak, of not using  variables in the primary models,
 this findng is counter intuitive and contradicts many crime studies
where factors beyond temperature have a clear impact on crime levels. 

I suspect, they were forced to test the control variables outside the
primary analysis because of different time intervals for the relevant
data. Often the socio-economic / LE data is available in annual form,
and the impact analysis -- crime and temp data was weekly. When we
locate a copy of the study, we will have a better idea.







------------------------ Yahoo! Groups Sponsor --------------------~--> 
Get fast access to your favorite Yahoo! Groups. Make Yahoo! your home page
http://us.click.yahoo.com/dpRU5A/wUILAA/yQLSAA/JjtolB/TM
--------------------------------------------------------------------~-> 

To subscribe, send a message to:
[EMAIL PROTECTED]

Or go to: 
http://groups.yahoo.com/group/FairfieldLife/
and click 'Join This Group!' 
Yahoo! Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/FairfieldLife/

<*> To unsubscribe from this group, send an email to:
    [EMAIL PROTECTED]

<*> Your use of Yahoo! Groups is subject to:
    http://docs.yahoo.com/info/terms/
 


Reply via email to