Yes in the sense that in the absence of any
other evidence I would have guessed the reverse.
But a guess is just a guess.
Can't you graph the data and think about them directly? If all
states superimposed looks too much like spaghetti
you can create a slide slow by something like this.
Suppose your states are 1 to 50 with value labels.
forval i = 1/50 {
tsline lnq5 lnmin1 lngap if state == `i', sub(`i')
more
}
Nick
[email protected]
Jeannette Wicks-Lim
> Thanks for clarifying your response. What I find puzzling is
> that the lngap is
> behaving relatively similarly to lnq5 rather than lnmin1,
> since lnmin1 is
> nonstationary. Isn't it unusual for a ratio of a
> nonstationary and stationary
> variable to be stationary?
*
* 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/
