It seems there's a problem with the decimal point in gretl's french version.
Let's try for example with the script below:
series x = normal()
series y =normal()
summary x --by=y
Gretl ends up with a message error:
x == -2,
The symbol ',' is not valid in this context
phone: (216) 22 99 73 44
I would like to know if there is any gretl version
available for Chromebook (Chrome OS): basically it means if there is any
gretl app created for it.
Scopri istella, il nuovo
motore per il web italiano.
Istella garantisce risultati di qualità e
la possibilità di condividere, in modo semplice e veloce, documenti,
immagini, audio e video.
Usa istella, vai su
Scopri istella, il nuovo motore per il web italiano.
Istella garantisce risultati di qualità e la possibilità di condividere, in modo semplice e veloce, documenti, immagini, audio e video.
Usa istella, vai su http://www.istella.it?wtk=amc138614816829636
Sorry in advance because I know that possibly it is not the best forum to
ask for this question, due that this is not a very Gretl specific
question. But I would know the expert opinion of Gretl list about this, if
My question is as follows: In my briefly research career in economics, I
had never heard about "effect size" question. Recently, I have studied a
very basic course in statistics at coursera, and when I studied a topic on
Null Hypothesis Significance Testing, I saw this question very interesting.
I'm currently reading the book of Mc.Closkey "The cult of statistical
significance" and I discovered a new world for me.
It appears to be a very important question in psychometric or other
experimental sciences. But not at the moment in econometrics (this is very
criticized in the Mc Closkey book).
Usually my job is related with multiple regression (mostly with temporary
series) and I uses Gretl for this purposes.
Normally showing my results I compute the b (understandardized) regression
coefficient with his corresponding t-value and his p-value. No more.
Lastly, due that I usually works with small samples sizes (normally 30 or
40 subjects/years) I indicates in my regression output a bootstrap
confident interval for b (understandardized) or a bootstrap p-value. Gretl
does this job greatly.
In the peer-review process nobody say me nothing about effect sizes of
My specific question is: somebody at the list knows what is the common
measure of effect size in multiple linear regressions?. In the topic of
this statistical course it is highligted taht beta (standardized regression
coefficient, that GRETL does not provide by default, but there is a GRETL
package to do it) or confidence intervals as measures of effect size, but
looking for internet I have no clear answer to my question. I found some
effect size measures for R2, but not for the regression coefficients.
Are in your opinion bootstrap intervals or p-values enough information for
effects size estimation?
Somebody knows some interesting literature addressed to econometrics in
this sense? GRETL Manual does not any reference to this "effect size"
Thanks in advance and really sorry for any inconvenience.
University of Alicante
How to forecast out of sample using moving average filter?
Prof. Carlos A. S. de Andrade
LAPEA - Laboratório de Pesquisa em Economia Aplicada e Engenharia de
Universidade Federal de Campina Grande.
Centro de Humanidades
Unidade Acadêmica de Economia
I am struggling with a simple issue:
I have a data set of monthly time series that spans 2000:01 to 2012:12
I want to run "pca" in a subset, from 2000:01 to 2011:12,
So I change the sample using "smpl ; 2011:12" command before running
"pca" command and save all the principal components.
But when I change back to full sample I can see the principal
components running through all my sample -- so I assume the "pca" runs in
the full dataset and does not respect the "smpl" command.
Is this the case? If so, how do I run "pca" in a subset of data?
Thanks a lot.
Dr. Paulo Gustavo Grahl, CFA
I am teaching an econometrics course where we are testing for
In GRETL the Breusch-Godrey test is in TESTS--Autocorrelation and you set
the number of lags.
The problem is that the Gretl BG test does not adjust of the number of
lags. For example, suppose we have a sample of 100 observations.
The BG test (using 4 lags of the residual) regresses the current residual
against the x variables and 4 lagged residuals. We can then use the LM
test statistics = (100- 4)*R^2.
The problem that I have with GRETL is that its test statistics uses T=100
in the auxiliary regression instead of the proper 96 (we use up 4
observations with the lags.)
Why does Gretl do this?
Dear users and higher beings,
I have been using gretl for a year, starting from GUI and switching to
scripts now. I have two questions (pertains to x64 version 1.9.14 on Win 7
and x32 1.9.14 on Debian x32):
1. For model pre-testing, a quick look at distributions and outlier search
is useful. However, I find it tedious to invoke qqplot or any other
visualization for every variable every time. I tried to create a loop for
12 variables, like
loop foreach i price..prefarea
---but all I got is 12 .plt files in the working directory. Gretl executed
the loop in batch mode, but... can the loop be forced to execute the
commands in interactive mode? What is desired is twelve windows popping up,
just like a single ``qqplot var'' in the console does, just 12 times? Worse
still, a one-line ``qqplot var'' script run from a script editor creates a
file, too, but what is needed is an immediate visual plot.
The manual and reference contain very few mentionings of interactive mode,
so any help would be appreciated!
2. The manual states that in boxplot diagrams, ``the `whiskers' extend to
the minimum and maximum values''. However, if so, there would be no
outliers! I checked the source code and found an implementation of
whiskers: uq and lq stand for quartiles, then the interquartile range is
defined as d: double d = limit * (plt->uq - plt->lq); double xlo = plt->lq
- d; double xhi = plt->uq + d. By default, the limit is 1.5. Is this a
misprint in the manual? Maybe it should be changed to the lines from the
gnuplot manual, which says, “by default the whiskers extend from each end
of the box for a range equal to 1.5 times the interquartile range?”
Looking forward to hearing from you soon.
---Andreï Victorovitch Kostyrka---
Department of Mathematical Economics and Econometrics
Higher School of Economics
I tried to save the recursive test statistics of the qlrtest using the
standard qlrtest command within a script but I fail to do so. Instead it
is, der default, saved as a *.plt file. Here is an output example:
Quandt likelihood ratio test for structural break at an unknown point,
with 15 percent trimming:
The maximum F(7, 106) = 5.13156 occurs at observation 1988:3
Asymptotic p-value = 0.0002343 for chi-square(7) = 35.921
Is there some way to do this?
when I replicate Example 11.3 of Wooldridge's advance book(the second
edition), i encounter a problem. To run the pooled OLS, I write
diff lare lfare_1 concen
ols d_lfare d_lfare_1 d_concen
then, I get the outputs, which is different from the results of the Book.
Model 1: OLS, using observations 3-4596 (n = 4594)
Dependent variable: d_lfare
Heteroskedasticity-robust standard errors, variant HC1
coefficient std. error t-ratio p-value
const 7.61258e-05 0.00427816 0.01779 0.9858
d_lfare_1 -0.114781 0.0104885 -10.94 1.56e-27 ***
d_concen -0.292149 0.0520106 -5.617 2.06e-08 ***
Mean dependent var 0.000068 S.D. dependent var 0.294573
Sum squared resid 386.0532 S.E. of regression 0.289981
R-squared 0.031358 Adjusted R-squared 0.030936
F(2, 4591) 70.98109 P-value(F) 4.37e-31
Log-likelihood -830.0114 Akaike criterion 1666.023
Schwarz criterion 1685.320 Hannan-Quinn 1672.815