Thanks!

Well I run script without "--Progressive" option. Unfortunately, it dropped my second variable (ECT1) because it contains only 0-s. However, I need both in the equation ECT1 and ECT2. Do you have any solution?

Here is the copy of results:

Current session: 2015-05-18 15:13

? T=$nobs

Replaced scalar T = 94

? loop i=1..T --quiet

> genr ECT1 = ECT*(time <= i)

> genr ECT2 = ECT-ECT1

> ols D_ARM ECT1 ECT2 D_FRN1 D_ARM1

> genr LF = $lnl

> store "C:\Users\Svanidze\Desktop\LF.gdt" LF

> endloop

Model 767: OLS, using observations 2006:12-2014:08 (T = 93)

Dependent variable: D_ARM

Omitted because all values were zero: ECT1

coefficient std. error t-ratio p-value

--------------------------------------------------------

ECT2 −0.281632 0.0562133 −5.010 2.70e-06 ***

D_FRN1 −0.0226867 0.103735 −0.2187 0.8274

D_ARM1 0.131557 0.0901396 1.459 0.1479

******************** I skip the rest **********************************************

Best,

Miranda

On Mon, May 18, 2015 at 3:03 PM, Riccardo (Jack) Lucchetti <r.lucchetti@univpm.it> wrote:

On Mon, 18 May 2015, Miranda Svanidze wrote:

To clarify my problem written in the previous email, I am adding results
from Gretl.

I dn't understand why you need the --progressive flag. Is the following example of any help to you?

<hansl>
set echo off
set messages off

open AWM.gdt --quiet
series y = ln(YER)
series c = ln(PCR)
series z = y - c
diff y c

T = $nobs
series ll = NA

loop i=1..T --quiet
series zp = z * (time < i)
series zm = z - zp
ols d_c 0 d_y zp(-1) zm(-1) --quiet
ll[i] = $lnl
endloop

plot ll
options time-series with-lines
end plot --output=display
</hansl>

-------------------------------------------------------
Riccardo (Jack) Lucchetti
Dipartimento di Scienze Economiche e Sociali (DiSES)

Università Politecnica delle Marche
(formerly known as Università di Ancona)

r.lucchetti@univpm.it
http://www2.econ.univpm.it/servizi/hpp/lucchetti
-------------------------------------------------------