On Fri, 06 May 2005 22:28:09 -0400, JDG wrote

> Why would one expect the business cycle to have periodicity?

Isn't that why we call things "cycles" -- because we observe them to be
periodic?  What are you asking?

> Another important flaw is that economic growth for a given year is 
> not an independent variable.   It is serially correlated to the 
> previous year's economic growth.    

Are you describing autocorrelation or something else?  

One big question would be to discover if there's an autocorrelation lag, which
could reveal that there's a stronger correlation to economic growth that lags
n months after a Republican administration.  It's being presented without the
dates, so I can't even easily look for lag.  And I wouldn't be surprised to
find that there isn't enough data to reach good conclusions.

Somebody gimme a correlogram.

I'm skeptical about this stuff, too, mostly because I work so much with time
series and correlations that I know how difficult it is to make good
statistical sense of such data.

Nick
_______________________________________________
http://www.mccmedia.com/mailman/listinfo/brin-l

Reply via email to