2:53 p.m.
Why is the tab for accessing the Robust options labelled HCCME? Only one of
the choices is HCCME (cross-section). Wouldn't it make more sense to label
it ROBUST since it determines the behaviour of the robust option for ts,
panel and GARCH?
Cheers,
Lee
--
Lee Adkins
Professor of Economics
lee.adkins(a)okstate.edu
learneconometrics.com
5:03 p.m.
New subject: "make check" errors
This is only to report thta I am obtaining these errors when executing
"make check" ('configure', 'make' and then 'make check')
with the source
code from current CVS:
ignacio@U001082:~/gretlcomp/gretl$ make check
make -C tests check
make[1]: se ingresa al directorio «/home/ignacio/gretlcomp/gretl/tests»
gcc -msse2 -g -O2 -Wall -W -Wno-unused-parameter -Wno-sign-compare
-I../lib/src -DSTANDALONE -I/usr/include/libxml2 -c nistcheck.c
nistcheck.c: In function 'main':
nistcheck.c:871: error: 'Z' undeclared (first use in this function)
nistcheck.c:871: error: (Each undeclared identifier is reported only once
nistcheck.c:871: error: for each function it appears in.)
nistcheck.c:872: warning: passing argument 3 of 'read_nist_file' from
incompatible pointer type
nistcheck.c:297: note: expected 'struct mp_results **' but argument is
of type 'struct DATASET **'
nistcheck.c:872: warning: passing argument 4 of 'read_nist_file' from
incompatible pointer type
nistcheck.c:297: note: expected 'int *' but argument is of type 'struct
mp_results **'
nistcheck.c:872: warning: passing argument 5 of 'read_nist_file' from
incompatible pointer type
nistcheck.c:297: note: expected 'int **' but argument is of type 'int *'
nistcheck.c:872: warning: passing argument 6 of 'read_nist_file' from
incompatible pointer type
nistcheck.c:297: note: expected 'struct PRN *' but argument is of type
'int **'
nistcheck.c:872: error: too many arguments to function 'read_nist_file'
nistcheck.c:878: warning: passing argument 3 of 'run_gretl_comparison'
from incompatible pointer type
nistcheck.c:704: note: expected 'struct mp_results *' but argument is of
type 'struct DATASET *'
nistcheck.c:878: warning: passing argument 4 of 'run_gretl_comparison'
from incompatible pointer type
nistcheck.c:704: note: expected 'int *' but argument is of type 'struct
mp_results *'
nistcheck.c:878: warning: passing argument 6 of 'run_gretl_comparison'
from incompatible pointer type
nistcheck.c:704: note: expected 'struct PRN *' but argument is of type
'int *'
nistcheck.c:878: error: too many arguments to function
'run_gretl_comparison'
nistcheck.c:881: warning: passing argument 2 of
'run_gretl_mp_comparison' from incompatible pointer type
nistcheck.c:573: note: expected 'struct mp_results *' but argument is of
type 'struct DATASET *'
nistcheck.c:881: warning: passing argument 3 of
'run_gretl_mp_comparison' makes integer from pointer without a cast
nistcheck.c:573: note: expected 'int' but argument is of type 'struct
mp_results *'
nistcheck.c:881: warning: passing argument 4 of
'run_gretl_mp_comparison' makes pointer from integer without a cast
nistcheck.c:573: note: expected 'const int *' but argument is of type
'int'
nistcheck.c:881: warning: passing argument 6 of
'run_gretl_mp_comparison' from incompatible pointer type
nistcheck.c:573: note: expected 'struct PRN *' but argument is of type
'int *'
nistcheck.c:881: error: too many arguments to function
'run_gretl_mp_comparison'
nistcheck.c:885: error: too many arguments to function 'destroy_dataset'
make[1]: *** [nistcheck.o] Error 1
make[1]: se sale del directorio «/home/ignacio/gretlcomp/gretl/tests»
make: *** [check] Error 2
--
*Ignacio Díaz-Emparanza*
Zuzendaria/Director
ignacio.diaz-emparanza(a)ehu.es <mailto:ignacio.diaz-emparanza@ehu.es>
94 6013732
*EKONOMIA APLIKATUA III SAILA (EKONOMETRIA ETA ESTATISTIKA)/
DEPARTAMENTO DE ECONOMÍA APLICADA III (ECONOMETRÍA Y ESTADÍSTICA)
UPV/EHU*
Avda. Lehendakari Aguirre, 83 | 48015 BILBAO
*T.: +34 946013740* | *F.: +34 946013754*
*www.ea3.ehu.es* <http://www.ea3.ehu.es>
7:11 p.m.
New subject: "make check" errors
On Wed, Jul 13, 2011 at 1:03 PM, Ignacio Diaz-Emparanza
<ignacio.diaz-emparanza(a)ehu.es> wrote:
This is only to report thta I am obtaining these errors when executing "make
check"
('configure', 'make' and then 'make check') with the source code
from current CVS [...]
Thanks, Ignacio. That's now fixed.
Allin
4:04 a.m.
On Wed, Jul 13, 2011 at 10:53 AM, Lee Adkins <lee.adkins(a)okstate.edu> wrote:
Why is the tab for accessing the Robust options labelled HCCME? Only
one of
the choices is HCCME (cross-section). Wouldn't it make more sense to label
it ROBUST since it determines the behaviour of the robust option for ts,
panel and GARCH?
Well, I believe that every option under that heading is literally an HCCME
(Heteroskedasticity-Consistent Covariance Matrix Estimator). The term is
not quite exact, agreed, since some of the estimators are also
autocorrelation-constant, and therefore HACCMEs.
My feeling is that the term "robust" is a bit vague. Admittedly, in
mainstream econometrics it's most associated with robust covariance
matrix estimation, but IMO it has as good or better reason to be
associated with robust parameter estimation (e.g. via LAD).
If brevity wasn't an issue, I'd label this tab "Robust Covariance Matrix
Estimation". I'm open to suggestions on an improved brief version.
Allin Cottrell
4923
days inactive
4924
days old
3 comments
3 participants
participants (3)
-
Cottrell, Allin -
Ignacio Diaz-Emparanza -
Lee Adkins