Is there a relogit (for Stata) post-estimation command, an analogue to
estat class for logit on Stata? I want to calculate the percentage of
cases correctly predicted, and couldn't see if there was a command to
do it nicely. I've looked in the archives and online already, so my
apologies if I've overlooked something obvious.
--
relogit mailing list served by Harvard-MIT Data Center
List Address: relogit(a)lists.gking.harvard.edu
Subscribe/Unsubscribe: http://lists.gking.harvard.edu/?info=relogit
I want to estimate a panel logistic regression with rare events.
the results that i found were no significant: i have to meke a mistake and i don't know on which step exactly.
i download ur software relogit for stata, but i think that i didn't succeed to install it. Otherwise , how can i use relogit ( where is the command on stata software?)
Cordially,
Msn Account : ons1402(a)laposte.net
On 10/13/2008 06:19 PM, El Gaied Ons wrote:
Hello Mr GARY,
Thanks for your answer.
I tryed to update relogit software but i found this when i run STATA:
STATA: C:\Program Files\Stata10\
UPDATES: C:\Program Files\Stata10\ado\updates\
BASE: C:\Program Files\Stata10\ado\base\
SITE: C:\Program Files\Stata10\ado\site\
PLUS: c:\ado\plus\
PERSONAL: c:\ado\personal\
OLDPLACE: c:\ado\
There is no "stbplus " as mentionned on the file "readme"
Please let me know , how can i proceed?
Cordially,
EL GAIED Ons
International Development/International Trade
Tél: 00216 71808049
Fax: 00216 71806962
----- Original Message -----
Hi,
I am using relogit in one of my papers. I run the same model with three different dependent variables. I works just fine with one of them. But, it doesn't work with the other two, which are no more than subsets of this first variable.
I get the following error messages:
1.
conformability error;
You have issued a matrix command attempting to combine two
matrices that are not conformable, for example, multiplying
a 3x2 matrix by a 3x3 matrix. You will also get this message
if you attempt an operation that requires a square matrix and
the matrix is not square.
However, I know it is a similar matrix with the one that works. The variables do not change, but the dependent one which also takes values 0 and 1 just like the one that works.
2.
matrix has missing values;
You have issued a matrix command attempting a matrix operation
that, were it carried out, would result in a matrix with missing
values; dividing by zero is a good example.
This is a little more perplexing. The model is the same. There is no missing data.
Any suggestions?
Ada Hyso, Doctoral Candidate
Department of Political Science
University of Wisconsin-Milwaukee
Milwaukee, WI 53201
Fax 414-229-5021
E-mail: adahyso(a)uwm.edu
--
relogit mailing list served by Harvard-MIT Data Center
List Address: relogit(a)lists.gking.harvard.edu
Subscribe/Unsubscribe: http://lists.gking.harvard.edu/?info=relogit
Dear Dr. King,
My co-authors and I were interested in implementing a latent class
analysis using rare events data and it seems the current relogit set-up
does not allow for this. We tried to then replicate the basic relogit
procedure in SAS but ran up against the limitations of SAS's matrix
inversion capabilities. I was wondering if there is a way to program the
rare events methodology using SAS. Are you aware of such an
implementation, or a procedure in relogit that may allow us to do this?
Can you please advise us?
Regards,
Deepa Chandrasekaran
--
Assistant Professor of Marketing
College of Business & Economics
Lehigh University
621 Taylor St., Bethlehem, PA 18015
Email: dec207(a)lehigh.edu
Phone (Office): 610-758-6582
Fax: 610-758-6941
--
relogit mailing list served by Harvard-MIT Data Center
List Address: relogit(a)lists.gking.harvard.edu
Subscribe/Unsubscribe: http://lists.gking.harvard.edu/?info=relogit
Dear Professor King,
Can I use relogit with Stata version other than Stata 6, let say Stata 9?
Thank you,
Ada Hyso
Doctoral Candidate
University of Wisconsin-Milwaukee
Department of Political Science
Milwaukee, WI 53201
Phone 414-229-2369/4221
Fax 414-229-5021
E-mail: adahyso(a)uwm.edu
--
relogit mailing list served by Harvard-MIT Data Center
List Address: relogit(a)lists.gking.harvard.edu
Subscribe/Unsubscribe: http://lists.gking.harvard.edu/?info=relogit
calculating R^2 and fit statistics in general doesn't make sense for
comparing relogit and logit. Logit will fit better, but relogit will be
less biased. In almost all applications its bias rather than fit that
we're shooting for. its the same for comparing OLS and WLS by the way:
WLS will never fit better than OLS (which after all is _least_ squares)
but if done properly WLS will have better statistical properties.
Best,
Gary
On Thu, 22 Mar 2007, Stijn Bruyneel wrote:
> Dear authors of RELOGIT,
>
>
> I am a Phd student who is currently investigating entrepreneurship. I make use
> of a dataset in which only few events are present. To investigate the impact of
> rare events on my logistic results, I make use of your RELOGIT package written
> for STATA.
>
> In my work I try to compare standard logistic results with the rare events
> results in order to evaluate the necessity of using the RELOGIT package. In
> order to make a good comparison, I would like to generate R² in the relogit
> regression. Does the RELOGIT package supply such modality?
>
> If this is not the case, and I should want to calculate R² myself, is there
> anything specific I need to take into account when accurately calculating R²?
>
> Thanks for your advice,
>
> Stijn Bruyneel
> Ph.D. student
> Hasselt University
> Department of Business Administration
> Faculty of Applied Economics
> Agoralaan - building D
> 3590 Diepenbeek (Belgium)
>
> tel: +32 11 26 86 38
> fax: +32 11 26 87 00
> e-mail: stijn.bruyneel(a)uhasselt.be
>
>
Dear all,
At the moment I am conducting some research on suicide, and I am
> trying to use Relogit. I am using the package
> for Stata, and I have a question.
>>
>
> First I estimate the model with an ordinary logit command, and then
> "keep if e(sample)". After that I try to use the relogit command using
> the exact same model, but I get the error message "matrix has missing
> values". I have gone through all the variables, and there are no
> missing values.
>
> Have you come across this problem before, and do you know what can
> cause it?
>
>
>
> Thank you in advance,
>
>
>
> Monica K Nordvik
>
Monica K Nordvik
PhD student
Department of Sociology
Stockholm University
S-106 91 Stockholm
Sweden
+46-8-16 17 22
monica.nordvik(a)sociology.su.se
Dear all,
I have used -relogit- to develop my model for an even that occurs in
0.97% of my sample. I now would like to compute the predicted
probabilities for my sample of 11,618. Normally I would type
predict p1
and then I could look at different graphs to see how well the model fit
the data, but when I type the above command in I get probabilities
greater than 1 and I don't know what they mean. I see that in setx you
can set it up to calculate probabilities given certain values of each
independent variable, which is really interesting, but I guess I was
looking for a way to get probabilities for everyone. Is there a way to
do this?
Any advice would be greatly appreciated.
Kind Regards,
Bellinda
--
relogit mailing list served by Harvard-MIT Data Center
List Address: relogit(a)lists.gking.harvard.edu
Subscribe/Unsubscribe: http://lists.gking.harvard.edu/?info=relogit
thanks for your note. pls see below...
On Wed, 27 Jul 2005, Parry Clarke wrote:
> Dear Professor King,
>
>
> My name is Parry Clarke and I am currently writing up my PhD on baboon
> intersexual conflict. Im sorry to bother you but I was wondering if you
> could answer some questions regarding Logistic Regression of Rare events.
>
> I am currently attempting to model the occurrence of male aggression
> directed at oestrous females. The behaviour itself is fairly rare and so
> I have binary coded its occurrence. In total I have 1745 sampling units,
> or rows of data, with only 21 of these coded 1 and 1724 coded 0.
> Initially I carried out standard logistic regression, but found that
> when it came to diagnosis all my influential points were my 1s. So
> that, if deletion diagnostics were performed I was left with no variance
> In addition, my intercepts did not seem very convincing.
>
> However, having now discovered your work on the subject and the package you
> have written for R things are looking up!, but I just have a couple of
> questions:
>
> 1)Are there diagnostics unique to the rare event analysis?
> 2)Is influential point examination redundant in rare event logistic
> regression?
> 3)How do I get deviance estimates of the final model?
relogit estimates the same coefficients as logit from the same model.
they do differ tho in order to get better properties. so just as using
weighted least squares will give different answers -- and will fit the
data less well than least squares -- we generally prefer wls to ls when
there are weights available. so in both relogit and wls, you have the
same issue of how to deal with diagnostics. there are no special
diagnostics, but in both (and lots of other methods) the issue is that you
can't really treat all the observations equally and an outlier for one
observation isn't the same as another. so in relogit for example, an
extra 1 inadvertently included in the dataset will be much more
consequential than an extra 0.
> 4)Part of my analysis has to been trying to relate other forms of male
> aggression to oestrous female-directed aggression. However, these other forms
> are also rare and so when I enter them into the analysis as a dichotomous
> explanatory variable I get answers that are not really supported by the
> observed data: For example: overlap between the 1s in the response and
> explanatory variables may only be 2 or 3 data points but the final model
> suggests that a large amount of the deviance is explained and the
> coefficients are highly significant: Is this simply an artifact of the rarity
> of both the dependent and independent variable and am I, therefore, better
> off excluding them from the multivariate analysis and doing separate
> contingency table analysis with them?
rare events in explanatory variables is a different issue involving the
sensitivity of the estimates to the coding of X. since almost all
relevant models are conditional on X, you have little choice but to run
the analysis as is unless you reconceptualize the project (such as having
both variables being dependent variables).
> 5)In a number of your articles (e.g. Explaining Rare Events In
> International Relations) your talk at length about sampling strategies
> and database trimming to create favourable ratio of 0s to 1s and to cut
> down on costs. In a situation such as mine where the data is collected
> and the database fixed is there any need to trim and subset. If so how
> do you suggest I go about that and is there a command in R for randomly
> selecting subsets of data?
if you have the data, you should use it. no reason to subsample at that
point. we do it to demonstrate what would happen if you couldn't afford
to collect all the data, as is the case in many fields. but more data are
better generally and here too.
>
> Once again I am sorry to bother you with such a lengthy email and I hope it
> is not too much of an imposition.
best of luck with your research,
Gary King
>
>
> Yours sincerely,
>
> Parry Clarke.
>
> _________________________________________________________________
> Winks & nudges are here - download MSN Messenger 7.0 today!
> http://messenger.msn.co.uk
>
>
On Wed, 13 Jul 2005, Barbaresco Gabriele wrote:
>> Dear Prof. King,
>>
>> I have read with great interest the following your papers on logit and
>> rare events:
>>
>> Logistic regression in rare events Explaining rare events in
>> international relations
>>
>> I really appreciated them, also due to the lack of academic literature
>> on this important topics. I am currently working on a model to forecast
>> companies failure and so I am dealing with logit and rare events.
>>
>> Referring to the second paper, the less technical one, I'd like to be
>> advised on the following points:
>>
>> 1) Pag. 697, I see that the standard errors of betas depend on the
>> estimated probability, that in rare events is usually very far from
>> 0.5 , but greater (if the model works well) for Y=1 than for Y=0.
>> So, you argue, the ones have more informative power. To reduce these
>> errors, can I balance the sample (50% of ones and 50% of zeros) or
>> better use the original sample balancing observations that they are
>> equally weighted? It seems from my simulation that rebalancing the
>> sample lead to better Wald test and also to better pseudo Rsquared
>> measures of fitting. In doing so the estimated probability range
>> from 0 to 1, and 0.5 could be used as cutoff.
if you can 'balance' the sample by collecting additional data (the 1s),
then that's great. but if you're talking about discarding 0s, that's not
generally a good idea (although there are possible exceptions unrelated to
your point). in statistics, more data are better.
>> 2) Page 704, here again it seems to me that the bias coefficient on the
>> costant is smaller the bigger is the sample (n stand at denominator)
>> and the closer the probability to 0.5. I work with a sample of
>> around 180.000 items, if I rebalance the sample as above, can I
>> reduce this source of bias too?
probably not.
>>
>> In a nutshell, it seems to me that rebalancing the sample can reduce
>> some of the problems you highlight, but it seem also that you never
>> suggest such a solution, so I wonder if I have misunderstood the
>> meaning of your paper.
we thought about it and tried it when writing the paper, and there is
probably still something going on that might be useful from a theoreetical
perspective, but from you perspective, I wouldn't discard any data if you
don't have to.
Gary King
---
Gary King
Institute for Quantitative Social Science
Harvard University, 34 Kirkland St, Cambridge, MA 02138
http://GKing.Harvard.Edu, King(a)Harvard.Edu
Direct 617-495-2027, Assistant 495-9271, eFax 812-8581
>>
>> Sorry for my boring questions, and thank you very much for your kind
>> assistance.
>>
>> Best regards,
>>
>> Mr. Gabriele Barbaresco
>> Research Dept. - Mediobanca
>> Milan - Italy
>>
>> www.mediobanca.it
>> www.mbres.it
>
-
relogit mailing list served by Harvard-MIT Data Center
List Address: relogit(a)latte.harvard.edu
Subscribe/Unsubscribe: http://lists.hmdc.harvard.edu/?info=relogit
