*
* Example from section 12.9.3 on pp 274-275
*
open data tablef2-2[1].txt
calendar 1960
data(format=prn,org=columns) 1960:1 1995:1 year g pg y pnc puc ppt pd pn ps pop
*
set loggpop = log(100*g/pop)
set logypop = log(y)
set logpg   = log(pg)
set logpnc  = log(pnc)
set logpuc  = log(puc)
set trend   = year
set predummy  = t<=1973:1
set postdummy = t>=1974:1
*
linreg loggpop / resids
# constant logpg logypop logpnc logpuc
@RegCorrs(number=5,method=yule)
*
* Use METHOD=PW on AR1 to do Prais-Winsten and METHOD=CORC for Cochrane-Orcutt.
* The results don't match because those in the text are done with a single
* iteration while RATS estimates to convergence.
*
ar1(method=pw) loggpop
# constant logpg logypop logpnc logpuc
ar1(method=corc) loggpop
# constant logpg logypop logpnc logpuc
*
* SEARCH is the maximum likelihood grid search procedure. Given the speed of
* current computers, it makes more sense to use this rather than iterative
* methods.
*
ar1(method=search) loggpop
# constant logpg logypop logpnc logpuc
*
linreg(define=ar2) resids
# resids{1 2}
filter(equation=ar2) loggpop / fgpop
filter(equation=ar2) logpg   / fpg
filter(equation=ar2) logypop / fypop
filter(equation=ar2) logpnc  / fpnc
filter(equation=ar2) logpuc  / fpuc
set fcons = 1-%beta(1)-%beta(2)
linreg(title="AR(2) regression") fgpop
# fcons fpg fypop fpnc fpuc