Hi,
I have two questions for Q3:
1. When I am using zelig (code specified in question) to estimate the model, am I supposed to estimate the model using the cleaned-out data (per instructions in lines before 1.1), or the original data? The code says "data = data", and I'm not quite sure what "data" should refer to.
2. Should I gauge the necessity of of rare events specification by some rule, like if the 95% CIs for both methods overlap, then there's not much to investigate, whereas if they do not overlap, then perhaps the rare events specification is worth investigating?
Thanks!
- Joseph
Hey guys,
Not to continue to pester people but when I run:
rare<-zelig(onset~warl+gdpenl+lpopl1+lmtnest+ncontig+Oil+nwstate+instab+
polity2l+ethfrac+relfrac, data=data, model="relogit")
i.e. the given code
I get the error msgs:
Error in family$linkfun(mustart) : Value 0 out of range (0, 1).
Does anyone see the error?
Sorry to bug everyone...I just can't actually do 3.1 if I can't get the
given code to load.
~Abby
On Wed, Mar 10, 2010 at 6:58 PM, Gavinlertvatana, Poj <
pgavinlertvatana at hbs.edu> wrote:
> Hi,
>
> I have two questions for Q3:
>
> 1. When I am using zelig (code specified in question) to estimate the
> model, am I supposed to estimate the model using the cleaned-out data (per
> instructions in lines before 1.1), or the original data? The code says
> "data = data", and I'm not quite sure what "data" should refer to.
>
> 2. Should I gauge the necessity of of rare events specification by some
> rule, like if the 95% CIs for both methods overlap, then there's not much to
> investigate, whereas if they do not overlap, then perhaps the rare events
> specification is worth investigating?
>
> Thanks!
> - Joseph
> _______________________________________________
> gov2001-l mailing list
> gov2001-l at lists.fas.harvard.edu
> http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
>
Hey guys,
I didn't actually mean the function we draw from...I meant to actually ask
should we be using plogis rather than pnorm in our actual loop, not which
function we should be drawing from.
Best,
Abby
Hey everyone,
I posted the tex code for section 6 so you can see an example of how the
beamer document class works. These section notes are relatively spare, but
there are a number of easy to implement beamer themes which can be found on
the web.
Best, Iain
Hi all,
In the example from section, we drew from "mvrorm" to create our sampling
distribution for beta...for 2.2 should we be drawing from this same
distribution or from a logistic distribution since we are working with
logistic regressions?
Best,
Abby
On Wed, Mar 10, 2010 at 4:14 PM, Michael Barnett <mlbarnett at gmail.com>wrote:
> Hey Ashley and all -
>
> I was a little wrong in my last answer - you're right that the
> baseline values of covariates is the same as in 2.1 except that we're
> setting nwstate to 0 or 1 and varying GDP. And the covariates we set
> in 2.1 are for countries at risk for civil war, so I guess they are
> high risk.
>
> -Michael
>
> On Mar 10, 2010, at 3:29 PM, Ashley Anderson wrote:
>
> > Hi all,
> >
> > So for problem 2.3, are we subsetting the data to include high risk
> > countries and then subsetting it again to include only those
> > countries that
> > are newly independent (or in the second case not independent) and
> > then doing
> > the confidence bands plots? I'm somewhat confused.
> >
> > -Ashley Anderson
> >
> >
> > _______________________________________________
> > gov2001-l mailing list
> > gov2001-l at lists.fas.harvard.edu
> > http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
>
> _______________________________________________
> gov2001-l mailing list
> gov2001-l at lists.fas.harvard.edu
> http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
>
Hi all,
So for problem 2.3, are we subsetting the data to include high risk
countries and then subsetting it again to include only those countries that
are newly independent (or in the second case not independent) and then doing
the confidence bands plots? I'm somewhat confused.
-Ashley Anderson
Thank you guys for your replies.
I was using the edit command, which wasn't allowing me to do what I wanted.
Instead, I went with a logical command, as Iain hinted, and that worked.
Thanks
Yousuf
On Wed, Mar 10, 2010 at 12:00 PM,
<gov2001-l-request at lists.fas.harvard.edu>wrote:
> Send gov2001-l mailing list submissions to
> gov2001-l at lists.fas.harvard.edu
>
> To subscribe or unsubscribe via the World Wide Web, visit
> http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
> or, via email, send a message with subject or body 'help' to
> gov2001-l-request at lists.fas.harvard.edu
>
> You can reach the person managing the list at
> gov2001-l-owner at lists.fas.harvard.edu
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of gov2001-l digest..."
>
>
> Today's Topics:
>
> 1. Calling the columns of the loaded dataset (Abdul Ahad Tariq)
> 2. Re: Calling the columns of the loaded dataset (Abigail Coplin)
> 3. problem editing onset variable (Yousuf)
> 4. different problem with data editing (Yousuf)
> 5. Re: different problem with data editing (Meryl Federman)
> 6. Re: different problem with data editing (Iain Osgood)
>
>
> ----------------------------------------------------------------------
>
> Message: 1
> Date: Tue, 9 Mar 2010 22:24:58 -0500
> From: Abdul Ahad Tariq <abdulahad.t at gmail.com>
> Subject: [gov2001] Calling the columns of the loaded dataset
> To: "Class List for Gov 2001/E-2001" <gov2001-l at lists.fas.harvard.edu>
> Message-ID:
> <f330d63b1003091924q7a4c618dq129b727ea3acbf31 at mail.gmail.com>
> Content-Type: text/plain; charset="iso-8859-1"
>
> So after loading "fearondata.dta" (using the read.dta command as someone
> pointed out earlier), what command can I use to call its columns and such?
> Does read.dta() load the State file into R as a matrix?
>
> Thanks!
>
Dear guys:
I was able to solve the command loading problem for editing; however,
another challenge looms. When I change the 4 into 1, as instructed, the
changed value doesn't stay the same. How do I tell R to save the new value
on my table?
Thanks
--
Yousuf
Dear guys:
I was doing the assignment and everything was looking fine. I used the
na.omit( ) to get rid of NAs and then used the edit command to get rid of
4s. When I initially typed in the edit command, a table showed up, which
was great. After that, my R crashed. When I restarted it, and went through
the same process, the command only generated a text format editable version
of the data. How do I get the table format again?
thanks
--
Yousuf