On Fri, 1 Jan 2016, oleg_komashko@ukr.net wrote:

> I think it's a good idea You can download it at 
> http://ricardo.ecn.wfu.edu/gretl/staging_fnfiles/lagreg.zip I'm not 
> sure, it seems less general but more easy to use It's intended mostly 
> for intermediate level courses Also I almost  finished functions for 
> automated mean and median lags computing Oleh
>
> 1 січня 2016, 13:39:01, від "Artur Tarassow" < artur.tarassow@googlemail.com >:
>
> Actually the dynMultipl-Package needs some revision. I already started 
> to update the package and hopefully will upload it soon.

None of my business, really, but if you two guys could cooperate for 
putting together a nice package for ARDL/ECM, that would be _very_ nice.

By the way: I've noticed that people tend to overlook the filter() 
function, which can do amazing things for you ;-)

<hansl>
open AWM --quiet

series y = ln(YER)
series c = ln(PCR)
series w = ln(WLN/PCD)

ols c const time y(0 to -4) w(0 to -2) c(-1 to -2)

ycoef = $coeff[3:7]
wcoef = $coeff[8:10]
ARcoef = $coeff[11:12]

scalar horizon = 20
matrix shock = ones(horizon + 1, 2)
shock[2:,1] = 0
multy = filter(shock, ycoef, ARcoef)
multw = filter(shock, wcoef, ARcoef)

sim_y = quantile(filter(1 | mnormal(horizon, 1000), ycoef, ARcoef)', 0.5)'

print multy multw sim_y
</hansl>

I'm sure you two guys need no pedantic explanation of the above

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