gretl on Android
by John C Frain
Sometime ago there was a discussion on this list about the feasibility of
running gretl on Android. I have been looking at some of the recent
developments on the Google play store. In particular the app GNURoot
provides a method to install and run Linux applications on an Android
device without needing to root the device. One then installs a Linux
rootfs. I installed the Debian Wheezy rootfs app from the play store. Using
this I have installed R, emacs, ess and gretl using standard apt-get. The R
inferior process works with emacs. As regards gretl gretlcli works as
expected but gretl_x11 can not start the display.
This was done on a Kobo Arc running Android 4.1.1. The author of these
apps has a good version of Octave and various octave packages available on
the play store. There is also a version of Maxima for Octave available.
Proof of what can be done? Personally I will not be selling my laptop or
desktop and moving to Android in the short run.
John C Frain
3 Aranleigh Park
Rathfarnham
Dublin 14
Ireland
www.tcd.ie/Economics/staff/frainj/home.html
mailto:frainj@tcd.ie
mailto:frainj@gmail.com
10 years, 11 months
forecast variance decomposition, style = stacked bars (changed)
by Tim Nall
Created forecast variance decomposition graph, style = stacked bars.
Edited title. When I pressed "Apply", the stacked bars styles
disappeared (became crosses, stars and boxes that mark the points that
wold have been used in a line graph).
10 years, 11 months
Computing R-Squared in panel Random Effects
by Ondřej Dvouletý
Dear all,
I am facing a problem with estimating panel regression with Random Effects.
I did not find any way how to persuade Gretl to compute R-Squared.
Could you give me advice? Thank you very much for replies.
*Ondřej Dvouletý*
Mobil.: +420 728 431 027
e-mail: ondradvoulety(a)gmail.com
Skype: ondradvoulety
[image: ICON] <http://cz.linkedin.com/in/ondrejdvoulety/>
Studeněves 98
273 79 TUŘANY U SLANÉHO
10 years, 11 months
Hansl programming performance
by Henrique Andrade
Dear Gretl Community,
I would like to know if the following "strategies" could improve my scripts
in terms of the execution time:
(1) Write/open gdt files is faster than csv files?
(2) Using Gretl's home directory (located at my hard drive) is faster than
use a network drive?
(3) Specify the object type could reduce the execution time? (e.g. "scalar
x = 2" instead of "x = 2")
(4) Short variables/lists names could reduce the execution time?
(5) Matrices are longer than series (I'm talking about computer memory
usage)?
Best regards,
Henrique Andrade
10 years, 11 months
MLE Estimation of ACDP Model
by Daniel Bencik
Dear all, I am trying to ML estimate a model from http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1884237 and the following code, with Rng being the variable of interest,
RANGE = Rng
scalar rng = 0.1
scalar err = 0.2
scalar gamma = 1
mle ll = 0.5*ln(theta) - theta*lambda + RANGE *(ln(RANGE)-1) - lngamma(RANGE+1) + theta*RANGE( 1 + ln(lambda/RANGE))
series lambda = mean(RANGE)
series lambda = c + rng*RANGE(-1) + err*lambda(-1)
params gamma c rng err
end mle --robust
returns "lag order: argument 2 should be scalar, is series. Data types not conformable for operation". As far as I undestand, the problem is the first element of LL, but have no clue how to correct it. Any hint is much appreciated.
Best,
Daniel
10 years, 11 months