2:05 p.m.
I have two sets of data that are not time series. I would like to fit
separate linear models (intercept and slope) to each set of data and
then test if the models are the same (intercepts and slopes are equal).
I understand that this may be done with the Chow test and that Gretl
offers this test. Can I do what I want here with the Chow test in Gretl?
Thanks.
Tom
9:23 p.m.
On Sat, 15 Nov 2008, Thomas La Bone wrote:
I have two sets of data that are not time series. I would like
to fit separate linear models (intercept and slope) to each set
of data and then test if the models are the same (intercepts and
slopes are equal). I understand that this may be done with the
Chow test and that Gretl offers this test. Can I do what I want
here with the Chow test in Gretl?
Yes, you can.
Allin Cottrell
10:26 p.m.
On Sat, 15 Nov 2008, Allin Cottrell wrote:
On Sat, 15 Nov 2008, Thomas La Bone wrote:
> I have two sets of data that are not time series. I would like
> to fit separate linear models (intercept and slope) to each set
> of data and then test if the models are the same (intercepts and
> slopes are equal). I understand that this may be done with the
> Chow test and that Gretl offers this test. Can I do what I want
> here with the Chow test in Gretl?
Yes, you can.
You can also do this "by hand". Here is an example which may be close to
what you need; it uses one of the example datasets supplied with gretl, in
which we test the hypothesis that the relationship between age and illness
is the same for men and women.
<script>
open rac3d
list X = const AGE
list Z = null
loop foreach i X
fem_$i = SEX*$i
Z += fem_$i
end loop
ols ILLNESS X Z
omit Z --quiet
</script>
Hope this helps,
Riccardo (Jack) Lucchetti
Dipartimento di Economia
Università Politecnica delle Marche
r.lucchetti(a)univpm.it
http://www.econ.univpm.it/lucchetti
10:17 a.m.
El Saturday 15 November 2008 15:05:12 Thomas La Bone escribió:
I have two sets of data that are not time series. I would like to
fit
separate linear models (intercept and slope) to each set of data and
then test if the models are the same (intercepts and slopes are equal).
I understand that this may be done with the Chow test and that Gretl
offers this test. Can I do what I want here with the Chow test in Gretl?
Thanks.
Tom
Assuming the gretl dataset contains the two sets ordered, the first set in the
first part of the sample and the second in the second part, and the variables
in the two sets have the same names, you only have to select /Model/Ols, in
the model window, /Tests/Chow test, and now select the observation in which
the second set starts.
As the gretl help says: "[the Chow command] Provides a test for the null
hypothesis of no structural break at the given split point. The procedure is
to create a dummy variable which equals 1 from the split point specified by
obs to the end of the sample, 0 otherwise, and also interaction terms between
this dummy and the original regressors. An augmented regression is run
including these terms.
By default an F statistic is calculated, taking the augmented regression as
the unrestricted model and the original as the restricted. But if the
original model used a robust estimator for the covariance matrix, the test
statistic is a Wald chi-square value based on a robust estimator of the
covariance matrix for the augmented regression."
--
Ignacio Diaz-Emparanza
DEPARTAMENTO DE ECONOMÍA APLICADA III (ECONOMETRÍA Y ESTADÍSTICA)
UPV/EHU
Avda. Lehendakari Aguirre, 83 | 48015 BILBAO
T.: +34 946013732 | F.: +34 946013754
www.et.bs.ehu.es
5921
days inactive
5923
days old
3 comments
4 participants
participants (4)
-
Allin Cottrell -
Ignacio Diaz-Emparanza -
Riccardo (Jack) Lucchetti -
Thomas La Bone