12 Sep
2006
12 Sep
'06
5:11 p.m.
Hi,
We've just released a new version of Zelig with some bug fixes in
setx, and for the multinomial logit model. This version is currently
available both from CRAN and from Gary's website, and is compatible
with the current R release (2.3.1, not 2.4.0 alpha). Please let us
know as you use Zelig, and find more bugs.
Best,
Olivia
-
Zelig Mailing List, served by Harvard-MIT Data Center
Send messages: zelig(a)lists.gking.harvard.edu
[un]subscribe Options: http://lists.gking.harvard.edu/?info=zelig
Zelig program information: http://gking.harvard.edu/zelig/
13 Sep
13 Sep
3:11 p.m.
On 9/12/06, Olivia Lau <olau(a)fas.harvard.edu> wrote:
We've just released a new version of Zelig with
some bug fixes in
setx, and for the multinomial logit model. This version is currently
available both from CRAN and from Gary's website, and is compatible
with the current R release (2.3.1, not 2.4.0 alpha). Please let us
know as you use Zelig, and find more bugs.
Thanks for the new release, which fixed the problem I was having with
multiply-imputed datasets with mlogit(); however, I've come across
another oddity: if I specify interactions using R syntax (x*y for x,
y, and x times y; x:y for x times y only) in the call to zelig(), the
interactions appear to be dropped - but if I use the non-standard
I(x*y), they work. It's not a huge deal, but in a 3-way interaction
I'd rather just type x*y*z and be done with it.
Also, does anyone know how to tell summary() that I don't want
scientific notation for p-values? - knowing that the effect of party
identification is significant at p < 2.68 * 10^-32 is nice, but I'd
report this as < .001 and I'd rather be able to tell at a glance where
something is in the 0.001-1 range. I've tried fiddling with
options(digits=x) and options(scipen=x) to no avail.
Thanks in advance,
Chris
--
Christopher N. Lawrence <clawren6(a)slu.edu>
Assistant Professor of Political Science (non-tenure-track)
Saint Louis University
109 Fitzgerald Hall
3500 Lindell Boulevard
St. Louis, Missouri 63103-1021
Website: http://www.cnlawrence.com/
-
Zelig Mailing List, served by Harvard-MIT Data Center
Send messages: zelig(a)lists.gking.harvard.edu
[un]subscribe Options: http://lists.gking.harvard.edu/?info=zelig
Zelig program information: http://gking.harvard.edu/zelig/
14 Sep
14 Sep
1:12 p.m.
Hi,
Thanks for pointing this out, Chris. We've released a new version
(2.6-5, available on Gary's website and soon on CRAN) that fixes the
way interaction terms are handled in multinomial logit, bivariate
probit, and bivariate logit. They should take the standard R syntax
of x*y = x + y + x:y now.
For the scientific notation problem, you can use round() as follows:
round(coef(summary(z.out)), 3)
and substitute
however many digits you want. Note that for values
that round to less than 0.001, though, this will return 0.000.
Best,
Olivia
On 9/13/06, Christopher N. Lawrence <c.n.lawrence(a)gmail.com> wrote:
Thanks for the new release, which fixed the problem I was having with
multiply-imputed datasets with mlogit(); however, I've come across
another oddity: if I specify interactions using R syntax (x*y for x,
y, and x times y; x:y for x times y only) in the call to zelig(), the
interactions appear to be dropped - but if I use the non-standard
I(x*y), they work. It's not a huge deal, but in a 3-way interaction
I'd rather just type x*y*z and be done with it.
Also, does anyone know how to tell summary() that I don't want
scientific notation for p-values? - knowing that the effect of party
identification is significant at p < 2.68 * 10^-32 is nice, but I'd
report this as < .001 and I'd rather be able to tell at a glance where
something is in the 0.001-1 range. I've tried fiddling with
options(digits=x) and options(scipen=x) to no avail.
Thanks in advance,
Chris
-
Zelig Mailing List, served by Harvard-MIT Data Center
Send messages: zelig(a)lists.gking.harvard.edu
[un]subscribe Options: http://lists.gking.harvard.edu/?info=zelig
Zelig program information: http://gking.harvard.edu/zelig/
1:56 p.m.
You may also try
print(summary(z.out), digits = 3)
Kosuke
-----------------------------------------------------
Kosuke Imai Office: Corwin Hall 041
Assistant Professor Phone: 609-258-6601
Department of Politics eFax: 973-556-1929
Princeton University Email: kimai(a)Princeton.Edu
Princeton, NJ 08544-1012 http://imai.princeton.edu
-----------------------------------------------------
On Thu, 14 Sep 2006, Olivia Lau wrote:
Hi,
Thanks for pointing this out, Chris. We've released a new version
(2.6-5, available on Gary's website and soon on CRAN) that fixes the
way interaction terms are handled in multinomial logit, bivariate
probit, and bivariate logit. They should take the standard R syntax
of x*y = x + y + x:y now.
For the scientific notation problem, you can use round() as follows:
-
Zelig Mailing List, served by Harvard-MIT Data Center
Send messages: zelig(a)lists.gking.harvard.edu
[un]subscribe Options: http://lists.gking.harvard.edu/?info=zelig
Zelig program information: http://gking.harvard.edu/zelig/
round(coef(summary(z.out)), 3)
and
substitute however many digits you want. Note that for values
that round to less than 0.001, though, this will return 0.000.
Best,
Olivia
On 9/13/06, Christopher N. Lawrence <c.n.lawrence(a)gmail.com> wrote:
Thanks for the new release, which fixed the problem I was having with
multiply-imputed datasets with mlogit(); however, I've come across
another oddity: if I specify interactions using R syntax (x*y for x,
y, and x times y; x:y for x times y only) in the call to zelig(), the
interactions appear to be dropped - but if I use the non-standard
I(x*y), they work. It's not a huge deal, but in a 3-way interaction
I'd rather just type x*y*z and be done with it.
Also, does anyone know how to tell summary() that I don't want
scientific notation for p-values? - knowing that the effect of party
identification is significant at p < 2.68 * 10^-32 is nice, but I'd
report this as < .001 and I'd rather be able to tell at a glance where
something is in the 0.001-1 range. I've tried fiddling with
options(digits=x) and options(scipen=x) to no avail.
Thanks in advance,
Chris
-
Zelig Mailing List, served by Harvard-MIT Data Center
Send messages: zelig(a)lists.gking.harvard.edu
[un]subscribe Options: http://lists.gking.harvard.edu/?info=zelig
Zelig program information: http://gking.harvard.edu/zelig/
6457
days inactive
6459
days old
3 comments
3 participants
participants (3)
-
Christopher N. Lawrence -
Kosuke Imai -
Olivia Lau