Although
syntax varname
is fine as a variant on
syntax varlist
-syntax- still populates the -varlist-
macro with whatever you supplied
as varname. No local macro -varname-
is defined by -syntax-.
Thus I guess that
qui dfuller `varname',
should be
qui dfuller `varlist',
A separate bug in your program is
that you allow -if- and -in- but
do nothing to let them take effect.
Various commands need to be made
... if `touse'
Another bug is
local i Lags
which should be
local i = `lags'
Another class of bugs is confusion between tstat
and `tstat', etc. And there are some typos.
I note tacit assumptions about what your variables
are called (DL*).
Here is a hack.
capture program drop adfperron
* NJC was here
*! adfperron v1 Nourry 03oct2007
program adfperron, rclass
version 9.2
syntax varname(ts) [if] [in] [,Lags(int -1)]
marksample touse
qui count if `touse'
if r(N) == 0 error 2000
local i = `lags'
qui dfuller `varlist' if `touse', lags(`i') trend regress
*/first regression with i lags/
local bstat = _b[DL`i']
local sestat = _se[DL`i']
local tstat = `bstat'/`sestat'
*/computation of the t-stat of the 8th lag/
qui while abs(`tstat')< 1.645 & `i' > 0 {
local i = `i' - 1
dfuller `varlist' if `touse', lags(`i') trend regress
local bstat = _b[DL`i']
local sestat = _se[DL`i']
local tstat = `bstat'/`sestat'
/* while the last lag is not significant, drop it and
do a new regression without this term */
}
dfuller `varlist' if `touse', lags(`i') trend regress
end
Nick
[email protected]
Myriam Nourry
I am trying to write a program to test unit root (ADF
> test) by using Perron's procedure to choose the
> maximum lag length included in the regression.
>
> The idea is to perform a first regression with 8 lags.
> If the absolute value of the t-statistic on the eighth
> lagged term is less than -1,645 (i.e., approximate 5%
> significance level in an asymptotic normal
> distribution), then this term is dropped and the
> procedure is repeated for the seventh lagged term. The
> procedure is stopped when the last lagged term is
> significant, in which case all lesser
> first-differenced lagged terms remain in the test, or
> zero lagged terms will result.
>
> Therefore, I wrote the following program (this is my
> first program...) but it doesn't work and I can't see
> the errors. So I need your help !
>
> capture program drop adfperron
> *! adfperron v1 Nourry 03oct2007
> program define adfperron, rclass
> version 9.2
> syntax varname(ts) [if] [in] [,Lags(int -1)]
> marksample touse
>
> local i Lags
> qui dfuller `varname', lags(`i') trend regress
> */first regression with i lags/
> local bstat = "_b[DL`i']"
> local sestat = "_se[DL`i']"
> gen tstat = `bstat'/`sestat'
> */computation of the t-stat of the 8th lag/
> while abs(`tstat')< abs(-1.645) & `i'>0 {
> local i=`i'-1
> qui dfuller `varname', lags(`i') trend regress
> qui replace bstat = "_b[DL`i']"
> qui replace sestat = "_se[DL`i']"
> qui replace tstat = `bstat'/`sestat'
> */while the last lag is not significant, drop it and
> do a new regression without this term/
> }
> dfuller `varname', lags(`i') trend regress
> end
>
> When I run this program on Stata, I obtain the
> following message : `invalid name r(198)
>
> I hope you can help me to solve this problem.
*
* 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/
