Gretl User's guide State Space modeling - clarification
by Alecos Papadopoulos
On page 349 of gretl's user guide the state space model is defined as
Later, when presenting the state-space GUI (p. 369) we read that it
supports models of the form
My only question is about the time indices on the state vector in the
transition equation.
In the first model (p. 349), the starting values for the state variables
(alpha) should be for time t, and will also affect the first time point
of the measurement equation. In the second formulation (p. 369), these
starting values should be for time (t-1), they should lead to alpha
values for time t through the transition equation alone, which will then
affect the first time point of the measurement equation.
Which of the two does gretl do?
--
Alecos Papadopoulos PhD
Affiliate Researcher
Dpt of Economics, Athens University of Economics and Business
Foundation for Economic and Industrial Research (IOBE)
web: alecospapadopoulos.wordpress.com/
ORCID:0000-0003-2441-4550
2 weeks, 6 days
Storing string-valued series in a database
by Artur T.
Hi all,
The first time, I play around with the option to store a dataset as a
gretl database. This works, as mentioned in the command reference, for
cross-sectional and time-series data but not panel (ok so far).
I tried to store a string-valued series. However, the string-values get
lost when importing again, only showing the numerical value mapping a
string to some discrete positive number.
I could not find any details in the manual nor the help text whether
this is actually supported.
I did the following:
<hansl>
open australia.gdt
series y = 1
stringify(y, defarray("foo"))
</hansl>
Then:
- via the menu File -> Export data -> Store series as gretl DB-format
(.bin).
- Delete series y in the active dataset
- Open the database by dragging the bin file from the file explorer into
the gretl main window.
- Import series y.
Thanks in advance.
Artur
3 weeks, 3 days
BayTool Package
by Brian Revell
Hi
I have tried using the BayTool package with some success (in terms of
getting output),
I have read (perhaps not well enough) the detailed paper by Luca Pedini.
However, I am puzzled why when simply using the linear model with
conjugate prior, as a default I get two sets of posterior mean parameter
and se estimates -NI-post m and I post-m and NI-Pst se and I-post se. .
Looking at the graphs of the Gaussian Posterior density estimates, they
appear to more closely centred on the I-pos -mean. Although the
difference s between the N & O posterior estimates is small, what is the
basis for selecting one rather than the other?
Second question - how can I retrieve the fitted values and residuals from
theB-Took Package. Hopefully it resides somewhere to be retirieved post
estimation. Or is iit necessary to run the input data through the equation
to generate it onself -and if so, which set of posterior mean estimates
does one use?
I expect these are straightforward issues for a Bayesian -but not for a
frequentist.
PS -the BayTool package does not seem to attach permanently to the URL
"model" list of options.
regards
Brian J Revell
Professor Emeritus (Agricultural Economics)
Tel: home 01952 728153 Mobile 07976 538712
Address: Orchard Croft, Vineyard Rd, Homer, Much Wenlock TF!3 6NG
Alt. Email bjrevell(a)harper-adams.ac.uk
3 weeks, 3 days
Bai-Perron Structural Breaks Model with GUI
by Brian Revell
I am attempting a simple task of identifying structural breaks in a single
time series trend.
There is a single dep var t=1994-2021, two variables in X viz. constant and
time
The following error message comes up
The variable X is read-only
*** error in function SB_GUI, line 7
> list X = !nelem(X) ? null : X
Some guidance would be welcome
.
Brian J Revell
3 weeks, 3 days
$error = 1
by Marcin Błażejowski
Hi,
two questions:
1. what $error=1 means?
2. and why 'errmsg(1)' returns empty string?
Marcin
--
Marcin Błażejowski
3 weeks, 4 days
translations for gretl 2024a
by Cottrell, Allin
Apologies to anyone who's getting this message for a second time, but
we're planning to release gretl 2024a soon, so this would be a good
time for translators to send in any updates they can.
Thanks, as always, to those who take the trouble to make gretl multilingual!
Allin
3 weeks, 6 days
Print the last n rows of the dataset
by Artur T.
Hi all,
printing the initial rows of a dataset is simple:
<hansl>
open denmark.gdt --quiet
print -o --range=:5
</hansl>
Printing the last n rows, however, takes another extra line
<hansl>
scalar n = $nobs-5
print -o --range=n:
</hansl>
as this does not work and fails:
<hansl>
print -o --range=($nobs-5):
</hansl>
Is my analysis correct or does there exist any other clever way?
Best
Artur
1 month, 2 weeks
BVAR package questions
by Dionysio Lalounta
Hi all,
I have a few questions about BVAR package .
At first how can I save the forecasted values of the depedent variables , in order to calculate RMSE etc?
The second question is about the inclusion of exogenous vars in the model .
I wrote the following script
clear --dataset
# read data available up to 2023:4
open dbnomics
data Eurostat/namq_10_gdp/Q.CLV05_MEUR.SCA.B1GQ.EA --name="YER"
data Eurostat/namq_10_gdp/Q.PD05_EUR.SCA.B1GQ.EA --name="YED"
data ECB/FM/Q.U2.EUR.RT.MM.EURIBOR1MD_.HSTA --name="STN"
data ECB/FM/Q.U2.EUR.4F.CY.OILBRNI.HSTA --name="POILU"
# BVAR with exogenous vars
list POILU1 =POILU # create a list for the exogenous variable
series yer=100*ldiff(YER)
series yed =100*ldiff(YED)
list dep = yer yed STN # create list for endogenous vars
lag =4
type ="fixed" # minnesota
exog = POILU1
bundle mod1 = BVAR_setup(dep, lag, type,_(exog=POILU1)) # bundle with exogenous variable
matrix m= zeros(1,4)
m={80,85,80,86} #future values for exogenous var
smpl 2000:1 2022:4
bundle res1 = BVAR_posterior(mod1,"all",_(iter = 10000, fcast_h = 4,fcast_exog=m)) # with exogenous var
When run the above script I get the following error
#ERROR MESSAGE
#error! Provide fcast_h obs of exog (as matrix, rows) for fcast!
Thanks in advance
Dionysio
1 month, 2 weeks
estimatios results
by Paolo Chirico
Hi,
when I estimate a model, the coefficients are saved in $coeff, the SEs
in $stderr, but t-ratio and pvalue not in $test and $pvalue.
If I run the modtest command after ols coomand I get values that do not
correspond to the requested ones.
How come?
Thanks,
Paolo
--
Paolo Chirico
RU e Prof.Agg. di Statistica Economica
Università del Piemonte Orientale
Dip. di Giurisprudenza e Scienze Politiche,
Economiche e Sociali (DIGSPES)
Alessandria, Italia
1 month, 3 weeks
New Release of "fdensity" Package - Version 0.6
by Artur T.
Hello Gretl community,
I am excited to announce the release of version 0.6 of the "fdensity"
package for Gretl. This package allows you to compute and draw
conditional densities of variable x for each value of the discrete
variable d, also known as the "factor."
The key features and changes in this release are as follows:
- Support for various plotting options: We have introduced additional
plotting options to enhance the visualization of the conditional
densities. You now have more flexibility to customize and present your
results effectively.
- Internal refactoring: We have made internal improvements and
optimizations to enhance the package's performance and maintainability.
These changes ensure a smoother user experience and better overall
functionality.
For more detailed information and to access the latest version of the
"fdensity" package, please visit the following link:
https://gretl.sourceforge.net/current_fnfiles/fdensity.gfn
We encourage you to explore the new features and improvements in version
0.6. Your feedback, suggestions, and bug reports are highly appreciated,
as they help us in further enhancing the package.
Thank you for your ongoing support and contribution to the Gretl community.
Best regards,
Artur
1 month, 3 weeks
