Dear Achim Zeileis, dear John Fox,
Thank you for your time! Both worked well.
lrtest(Restrict, Full)
#Df LogLik Df Chisq Pr(>Chisq)
1 27 -882.00
2 28 -866.39 1 31.212 2.313e-08 ***
anova(Restrict, Full)
Resid. df Resid. Dev TestDf LR stat. Pr
Dear Faradj Koliev,
There is an anova() method for "polr" objects that computes LR chisquare tests
for nested models, so a short answer to your question is anova(Full,
Restricted).
The question, however, seems to reflect some misunderstandings. First aov()
fits linear analysis-of-variance mode
On Wed, 27 Jul 2016, Faradj Koliev wrote:
Dear all,
A quick question: Let?s say I have a full and a restricted model that looks something like this:
Full<- polr(Y ~ X1+X2+X3+X4, data=data, Hess = TRUE, method="logistic?) # ordered logistic regression
Restricted<- polr(Y ~ X1+X2+X3, data=da
Amanda Li uchicago.edu> writes:
>
> Hello,
>
> I am so sorry, but I have been struggling with
> the code for the entire day.
>
> I have a very simple dataset that looks like this:
> response=c(45,47,24,35,47,56,29)
> sub=c("A","A","B","B","C","C","C"£©
> time=c(1,2,1,2,1,2,3)
> gdata=cbind(re
> I have run a regression and want to calculate the likelihood
> of obtaining the sample.
> Is there a way in which I can use R to get this likelihood value?
See ?logLik
And see also ?help.search and ??. You would have found the above by typing
??likelihood at the command line in R
S Ellison
On Nov 10, 2012, at 11:57 AM, bgnumis bgnum wrote:
> Hi All
>
>
> I have to run multiple stimations and to compute Likelihhod ratio.
>
> If I compute ls function with coef and summary I can extract "outputs" that
> I need.
Do you mean "lm"?
>
> I am not able to find something similar to lo
Thanks for pointing that out. Made some modifications and it worked.
--
View this message in context:
http://r.789695.n4.nabble.com/likelihood-function-involving-integration-error-in-nlm-tp4646697p4646764.html
Sent from the R help mailing list archive at Nabble.com.
___
On 19-10-2012, at 04:40, stats12 wrote:
> Dear R users,
>
> I am trying to find the mle that involves integration.
>
> I am using the following code and get an error when I use the nlm function
>
> d<-matrix(c(1,1,0,0,0,0,0,0,2,1,0,0,1,1,0,1,2,2,1,0),nrow=10,ncol=2)
> h<-matrix(runif(20,0,1),
On Sun, 12 Jun 2011, Jorge Ivan Velez wrote:
Hi Diviya,
Take a look at the lrtest function in the lmtest package:
install.packages('lmtest)
require(lmtest)
?lrtest
Yes, when you have to nls() fits, say m1 and m2, you can do
lrtest(m1, m2)
However, I don't think that both m1 and m2 can be i
Hi Diviya,
Take a look at the lrtest function in the lmtest package:
install.packages('lmtest)
require(lmtest)
?lrtest
HTH,
Jorge
On Sun, Jun 12, 2011 at 1:16 PM, Diviya Smith <> wrote:
> Hello there,
>
> I want to perform a likelihood ratio test to check if a single exponential
> or a sum of
karuna m yahoo.com> writes:
> Can anybody tell me which R package has Lo-Mendell Rubin LR test and
> Bootstrap
> LR test to compare the model fit between k class and k+1 class model
> for Latent class analysis?
I don't know, but
library("sos")
findFn("Lo-Mendell")
findFn("{latent class ana
On Mar 25, 2011, at 12:17 PM, Michael Hecker wrote:
Hi,
I have a dataset of 78.903 news articles pertaining to 340 corporate
takeovers.
Mean 231.3871 [articles per takeover]
Std. Dev. 673.6395
I would like to calculate the probability of a certain number of
news articles if I had more tak
On 2010-04-30 12:42, jh556 wrote:
Some quick googling suggests that they are the same thing. Thanks for the
help!
And note that profile likelihood CIs are produced by default on
glm objects, i.e. R uses MASS's confint.glm for glm objects.
confint.default(your model) let's you compare with Wa
Some quick googling suggests that they are the same thing. Thanks for the
help!
--
View this message in context:
http://r.789695.n4.nabble.com/Likelihood-ratio-based-confidence-intervals-for-logistic-regression-tp2077303p2077354.html
Sent from the R help mailing list archive at Nabble.com.
___
jh556 wrote:
I'm applying logistic regression to a moderate sized data set for which I
believe Wald based confidence intervals on B coefficients are too
conservative. Some of the literature recommends using confidence intervals
based on the likelihood ratio in such cases, but I'm having diffic
Davnah Urbach dartmouth.edu> writes:
> Thanks for this answer but does that mean that working
> with the deviances is better? Or how else could I
> evaluate the importance of my random terms?
You should probably (a) search the archives of the
r-sig-mixed-models mailing list
and (b) ask this
>
Thanks for this answer but does that mean that working with the deviances is
better? Or how else could I evaluate the importance of my random terms?
Many thanks,
Davnah
> On Mar 14, 2010, at 8:12 PM, hadley wickham wrote:
>
>>> Based on a discussion found on the R mailing list but dating
> Based on a discussion found on the R mailing list but dating back to 2008, I
> have compared the log-likelihoods of the glm model and of the glmer model as
> follows:
>
> lrt <- function (obj1, obj2){
> L0 <- logLik(obj1)
> L1 <- logLik(obj2)
> L01 <- as.vector(- 2 * (L0 - L1))
> df <- attr(L1,
Jim Silverton wrote:
Is there any package available in R to do the following hypothesis tests?
Testing the means of two Poissons (equivalent to the difference of two
proportions)
Testing the equality of two proportions from binomials
Testing the equality of proprtions of two negative binomials
(
Is there any package available in R to do the following hypothesis tests?
Testing the means of two Poissons (equivalent to the difference of two
proportions)
Testing the equality of two proportions from binomials
Testing the equality of proprtions of two negative binomials
(both conditional and un
Thanks, I had a look at mlogit. It seems it does fit a multinomial logit
regression but - just as nnet or VGAM are doing it - it has a function that
tells you the fitted value, not the value that you have with a set of
parameters (which might not be the optimal ones). Or am I wrong on this?
Rong
Thanks for the book suggestion. I'll check it out tomorrow when the library
opens up.
Yes, it is a multilevel model, but its likelihood function is the sum of the
likelihood functions for the individual levels (i.e. a simple multinomial
logits) and some other terms (the priors). It is, essential
You may refer to mlogit for the ordinary multinomial regression. As
fas as I know, there are no functions for multilevel multinomial
regression.
Ronggui
2009/8/2 nikolay12 :
>
> Hi,
>
> I would like to apply the L-BFGS optimization algorithm to compute the MLE
> of a multilevel multinomial Logist
Hi: John Fox's CAR book has some very nice examples of how the multinomial
likelihood is estimated computationally speaking. But you mentioned
multilevel earlier which sounds more complex ?
On Aug 2, 2009, nikolay12 wrote:
Thanks a lot. The info about computing the gradient
Thanks a lot. The info about computing the gradient will be helpful.
I admit that I am somewhat confused about the likelihood function itself. It
is often said that you need to set a reference category. However, I found
two different implementations in Matlab for which setting the reference
categ
Hi,
Providing the gradient function is generally a good idea in optimization;
however, it is not necessary. Almost all optimization routines will compute
this using a simple finite-difference approximation, if they are not
user-specified. If your function is very complicated, then you are more
Joris,
Ridge regression is a type of regularized estimation approach. The objective
function for least-squares, (Y - Xb)^t (Y - Xb) is modified by adding a
quadratic penalty, k b^t b. Because of this the log-likelihood value (sum of
squared residuals), for a fixed k, does not have much meanin
For the third one:
?anova.glm
test=Chisq will be LRT.
For the first two, you can have the answer from ordinary stat book.
On Wed, Nov 5, 2008 at 1:11 PM, Maithili Shiva <[EMAIL PROTECTED]> wrote:
> Hi!
>
> I am working on the Logistic Regression using R. My R script is as follows
>
>
> ONS <-
Christophe LOOTS ifremer.fr> writes:
>
> Thank you so much for your help.
>
> The function "dbinom" seems to work very well.
>
> However, I'm a bit lost with the "dnorm" function.
>
> Apparently, I have to compute the mean "mu" and the standard deviation
> "sd" but what does it mean exactly?
This particular case with a random intercept model can be handled by
glmmML, by bootstrapping the p-value.
Best, Göran
On Thu, Jul 17, 2008 at 1:29 PM, Douglas Bates <[EMAIL PROTECTED]> wrote:
> On Thu, Jul 17, 2008 at 2:50 AM, Rune Haubo <[EMAIL PROTECTED]> wrote:
>> 2008/7/16 Dimitris Rizopoulo
On Thu, Jul 17, 2008 at 2:50 AM, Rune Haubo <[EMAIL PROTECTED]> wrote:
> 2008/7/16 Dimitris Rizopoulos <[EMAIL PROTECTED]>:
>> well, for computing the p-value you need to use pchisq() and dchisq() (check
>> ?dchisq for more info). For model fits with a logLik method you can directly
>> use the foll
2008/7/16 Dimitris Rizopoulos <[EMAIL PROTECTED]>:
> well, for computing the p-value you need to use pchisq() and dchisq() (check
> ?dchisq for more info). For model fits with a logLik method you can directly
> use the following simple function:
>
> lrt <- function (obj1, obj2) {
>L0 <- logLik(
well, for computing the p-value you need to use pchisq() and dchisq()
(check ?dchisq for more info). For model fits with a logLik method you
can directly use the following simple function:
lrt <- function (obj1, obj2) {
L0 <- logLik(obj1)
L1 <- logLik(obj2)
L01 <- as.vector(- 2 *
Christophe LOOTS ifremer.fr> writes:
>
> Hi,
>
> I've two fitted models, one binomial model with presence-absence data
> that predicts probability of presence and one gaussian model (normal or
> log-normal abundances).
>
> I would like to evaluate these models not on their capability of
> a
Dear List,
Probably i am missing something important in optimize:
llk.1st <- function(alpha){
x <- c(20.0, 23.9, 20.9, 23.8, 25.0, 24.0, 21.7, 23.8, 22.8, 23.1, 23.1, 23.5,
23.0, 23.0)
n <- length(x)
llk1 <- -n*log(gamma(alpha)) - n*alpha*log(sum(x)/(n*alpha)) + (alpha -
1)*(sum(log(x))) - (su
You asked for a hint.
> library(MASS)
> x <- c(20.0, 23.9, 20.9, 23.8, 25.0, 24.0,
21.7, 23.8, 22.8, 23.1, 23.1, 23.5,
23.0, 23.0)
> fitdistr(x, "gamma")
shape rate
316.56387213.780766
(119.585534) ( 5.209952)
To do it with more general and e
Dear David,
no, distrTEst won't help. It has a different intention.
We are currently working on a new package "distrMod"
(cf. https://r-forge.r-project.org/projects/distrmod/)
which sometime might have such a functionality.
Best,
Matthias
David Bickel wrote:
> Is there any automatic mechanism f
David Bickel wrote:
> Is there any automatic mechanism for extracting a likelihood or test
> statistic distribution (PDF or CDF) from an object of class "htest" or
> from another object of a general class encoding a hypothesis test
> result?
>
> I would like to have a function that takes "x", an ob
You could create an S3 generic that does it. That is not initially
any less work than the if statement but if you add new distribution
no existing code need be modified. Just add a new method for each
distribution to be supported:
getDistr <- function(x) {
.Class <- names(x$value$statist
On Sat, 5 Jan 2008, xinyi lin wrote:
> Hi,
>
> I want to do a global likelihood ratio test for the proportional odds
> logistic regression model and am unsure how to go about it. I am using
> the polr() function in library(MASS).
>
> 1. Is the p-value from the likelihood ratio test obtained by
> a
On Fri, 21 Sep 2007, Wensui Liu wrote:
> chris,
> as long as you know the log likelihood functions and the # of
> parameters in both models, a pencil and a piece of paper should be
> enough to calculate LR test.
True enough for the LR statistic.
Or follow the instructions in the _posting guide_
chris,
as long as you know the log likelihood functions and the # of
parameters in both models, a pencil and a piece of paper should be
enough to calculate LR test.
On 9/21/07, Chris Elsaesser <[EMAIL PROTECTED]> wrote:
> I would like to try a likelihood ratio test in place of waldtest.
> Ideally
Chris Elsaesser wrote:
> I would like to try a likelihood ratio test in place of waldtest.
> Ideally I'd like to provide two glm models, the second a submodel of the
> first, in the style of lrt
> (http://www.pik-potsdam.de/~hrust/tools/farismahelp/lrt.html). [lrt
> takes farimsa objects]
>
> Does
43 matches
Mail list logo
