extra i in formula in guide?
by Sven Schreiber
Hi Jack (I guess this goes to you...),
in the Fourier transform definition in section 6.5 of the guide,
am I correct in thinking that there is one i too many? I mean, plugging in
the definition of $\omega(j,k) = 2 \pi i jk/n$ into the formula which contains
$e^{-i\omega(j,k)}$, we get two i instead of one, right?
thanks,
sven
12 years
var --lagselect
by Marcin Błażejowski
Hi,
I noticed when we run 'var lag --lagselect' neither "--silent" nor
"--quiet" options have impact on what is printed.
IMHO since we have $test accessor option "--silent" would be useful.
Best Regards,
Marcin
--
Marcin Błażejowski
GG: 203127
12 years
VECM alpha standard errors not shown
by Sven Schreiber
Good morning,
just a cosmetic issue: I have just noticed that the standard errors of
the alpha coefficients are not shown in the standard VECM output. (I
know it's redundant since the ECTs are in the equations, but then
displaying alpha itself is also redundant, strictly speaking.)
In contrast, the standard errors *are* shown if I implement the
identifying restrictions (on beta for example) manually. It would be
nice if gretl does it by default in the standard setup, too.
thanks,
sven
12 years, 1 month
Gretl crashes
by Marcin Błażejowski
Hi,
I found a bug in Gretl.
When I've tried to generate 'fcast --out-of-sample --no-stats --quiet'
based on arima model I got crash and following GDB results:
---
Program received signal SIGSEGV, Segmentation fault.
create_Xb_series (Z=0xfd8460, xlist=0xf87d50, beta=0x0, pmod=0xee47c0,
fc=<optimized out>)
at ../lib/src/forecast.c:1273
1273 x = arima_difference_obs(Z[vi], t, delta, k);
---
Note, that it happens whene I generate forecasts inside the function
package. In normal script it works fine:
---
open bjg
smpl 1949:01 1959:12
arima 0 1 1 ; 0 1 1 ; lg --quiet
fcast --out-of-sample --no-stats
---
I use current CVS.
Regards,
Marcin
--
Marcin Błażejowski
GG: 203127
12 years, 1 month
multiple gretl instances once again
by Allin Cottrell
Despite what I said a couple of weeks ago, I think I've now found a
reasonably reliable way of showing sequence numbers in the gretl
title bar when the user has two or more instances (copies) of gretl
running simultaneously.
Here's what's supposed to happen. The first gretl instance is always
just titled "gretl". The next concurrent instance is titled "gretl
(2)", the next again, "gretl (3)", and so on.
There's an ambiguity over what happens if, say, you have "gretl" and
"gretl (2)" running, then you close the first instance, then you
start another. The ambiguity is resolved in this way: the new
instance will have a sequence number 1 higher than the
highest-numbered running instance, so in the case just indicated it
will appear as "gretl (3)".
I believe this is now working on all OSes, but it would be nice to
have that confirmed (using CVS/snapshots, of course).
Allin
12 years, 1 month
