Questions tagged [quasi-likelihood]

In GLMs, quasi-likelihood estimation is a way to allow over- or under-dispersion by choosing an appropriate variance function.

Filter by
Sorted by
Tagged with
2
votes
0answers
14 views

How to compute the sandwich variance ML estimator in R

I'm currently estimating a DCC-type model by maximum likelihood. Im using the command solnp and it return an object where I can compute the Hessian H evaluated at ...
0
votes
1answer
30 views

Test for significant differences for data between 0 and 1

I have to test for significant differences between scenarios. Data consist of the length of a segment divided by the total length of the network. They are distributed between 0 (never equal to 0) and ...
0
votes
1answer
50 views

Output in two-part Fractional Response Model (FRM) package incorrect?

I was experimenting with the Fractional Response Model (FRM) package, and decided to replicate the results using the base GLM package to better understand the theory. I am able to replicate the ...
2
votes
1answer
56 views

Quasi-likelihood function

I got stuck in the derivation of the quasi-likelihood function. Namely, given an i.i.d sample $\{Y_i,X_i \}_{i=1}^n$ with $n$ the sample size, let the conditional mean and variance functions be ...
0
votes
0answers
23 views

Comparing the marginal effect of a GLM to the OLS estimates

My question is, whether there is any way to (somewhat) compare the marginal effect of a GLM estimate to an OLS estimate. As in, "since the OLS and GLM results are very similar, I will favour OLS ...
0
votes
1answer
36 views

Comparing the marginal effects of glm output to polr output

I have a dependent variable that is technically ordinal, so I ran a ordered probit model (polr). However, an ordered probit model does not produce any residuals ...
1
vote
0answers
56 views

How flexible is Stata's ivpois? Could I use it for a (quasi) binomial distribution?

According to this post on statalist, Stata's ivpois (an instrumental variable approach) is pretty flexible, with very little assumptions. The problem mentioned in ...
1
vote
0answers
71 views

How to do a Control Function (CF) / Two Stage Residual Inclusion (2SRI) with an ordinal dependent variable in the first stage and a glm in the second

I am trying to use a Control Function (CF) / Two Stage Residual Inclusion (2SRI) approach, because the modeled relationship that I am trying to estimate is non-linear (my dependent variable has a ...
1
vote
2answers
44 views

Getting the (Stata) margins from fractional regression (=glm with family quasibinomial) for an ordinal variable in R

I first found this really nice Stata video on fractional regression (the dependent variable is a proportion including 0 and 1). I am especially interested in how he applies the margin approach to ...
3
votes
2answers
154 views

What is the interpretation of a glm coefficient on a dependent variable that has a % interpretation

I have a dependent variable that takes on values between 0 and 1, including 0 and 1. The variable signifies a proportion (0 = nothing, 1 = all). I am running a model of the type: ...
0
votes
0answers
26 views

Intuition for reasoning between Quasi-poisson and Negative Binomial regression

I am aware of the several similar questions existing here (like this, this or this), but my question is slightly different and remains after going through most of those posts. Specifically, my ...
0
votes
1answer
24 views

Changes in significance between Poisson & quasipoisson glm

I am fairly new to GLMs, and am currently practising and testing with an insurance dataset, after many tries, I am modeling the frequency (counting model of the number of claims) and I have several ...
0
votes
0answers
20 views

A problem of likelihood function in a dynamic setting?

I'm having a problem regarding perhaps conditional maximum likelihood problem, but I'm not sure. Suppose time horizon we consider is $T=4$, our goal is to minimize the loss function $$ \sum_{t=2}^T L(...
1
vote
0answers
15 views

Relationship between quasi-likelihood and “GLM-conjugate” models?

Suppose we have a response variable that represents proportions with a poorly defined denominator. Two ways to handle this (1) a quasibinomial model, which assumes only that the variance is ...
0
votes
0answers
9 views

Do we have to check heteroscedasticity when estimating quasibinomial model?

As we all know we are not interested in checking heteroscedasticity when estimating logit model when our variable is binary (taking only values 0 and 1). However do we also don't have to test ...
0
votes
0answers
14 views

Is there any sense of applying cross validation to quasibinomial variable?

We can apply without any doubts cross validation for linear models as well as for binary models. For linear models for example we can output RMSE and MAE and Accuracy for binary models. But I have ...
0
votes
0answers
14 views

GLM on features selected by PLS-DA

First question: Can we use GLM on specific variables selected by PLS-DA latent components? To obtein p-value of response of prediction (e.g.for each selecetd variable of comp 1)? Second question: What ...
1
vote
1answer
54 views

Comparing model efficiency

I hope you all don't mind me asking this question. I have two models : general linear mixed effects model ...
4
votes
0answers
99 views

Why is my quasibinomial GLM estimator biased - Monte Carlo simulation

I'm playing with some Monte Carlo simulations to get an idea of the properties of some linear and non-linear models. The linear OLS model in my case is specified as: $Y_t = \beta_0 + \beta_1x+ \...
1
vote
0answers
24 views

If my data is overdispersed when binomial but not when Poisson does that alone mean I should use Poisson GLM?

Top is Poisson, bottom is Binomial Source of variation is different people observed over a week long period. 118 people to be exact. want to make predictions about the proportion of days/week one ...
1
vote
0answers
38 views

Avoiding data dredging with quasibinomial GLMs

Because my data is a bit overdispersed I am using quasibinomial GLM to analyse them however this means I cannot use AICs to compare my models. I am therefore using drop1 and update functions to do ...
0
votes
0answers
20 views

Should “kenward-roger”/“satterthwaite” degrees of freedom be used with “GLMM” fitted with penalized quasilikelihood"? [duplicate]

Should kenward-roger degrees of freedom correction be used with generalized linear mixed models fitted with penalized quasilikelihood/"PQL"? There's no reason it isn't implementable but the ...
1
vote
0answers
38 views

Equivalence between prediction interval and hypothesis test on Poisson predicted values with 2/3 power transformation

I fitted a quasi-Poisson regression model to predict the value of a certain counting variable. Once the predicted value $\widehat{\mu_{0}}$ was obtained, I calculated the upper limit of its prediction ...
1
vote
0answers
30 views

Overdispersion in logistic model

I'm relatively a newbie in R, and I've been trying to make a silly example of logistic regression to predict, according to Age and Sex whether someone dies of corona or not. I'm from Colombia, so my ...
0
votes
1answer
187 views

Interpreting coefficents on quasibinomial model

This might seem like a pretty basic question but I've scoured seemingly everywhere and can't get a definitive answer. I have a response parameter "rr" bound by 0-1 which is essentially the ...
0
votes
1answer
128 views

MASS::glmmPQL diagnostic

I am fitting models with MASS::glmmPQL of the form ...
2
votes
1answer
457 views

Quasi Poisson vs Negative Binomial [duplicate]

I read in several sources that the Quasi Poisson model and the Negative Binomial, should produce (on average) the same results. I tried a simple example and, although very close to each other, the ...
0
votes
0answers
34 views

R binomial model where DV is a proportion and distribution appears Bimodal

I've been attempting to fit a binomial model to a data set of 1,000,000 accounts where the DV (rr) is a percentage of account balance that been has paid (EX account with total owed of 100 dollars has ...
0
votes
0answers
41 views

Why can we use a deviance test for a quasi likelihood models?

I know that the reason why you don't obtain a value for AIC if you fit a quasi-likelihood model is because a quasi-likelihood is not a real likelihood, and to obtain a value for AIC, you need a real ...
1
vote
0answers
119 views

Testing for endogeneity in a negative binomial model

I'm trying to fit a negative binomial model to my data because the dependent variable exhibits overdispersion. However, one of my reviewers is insisting that I also test for endogeneity. He or she is ...
0
votes
1answer
139 views

Interpreting interaction among 2 categorical IV in quasi-poisson regression

In my dataset, I'm looking at the impacts of developmental and immune phenotypes on morbidity- specifically, I want to determine if developmental phenotype has an effect on the difference in morbidity ...
1
vote
0answers
20 views

Can we mix conlcusions from Poisson and Quasi-Poisson?

Currently I'm working with ecological studies, where my response is a count variable. I need to estimate several models, each one represents a city. Afterwards I aggregate them to obtain meta-analysis ...
1
vote
1answer
142 views

Whats the difference between logistic regression and fractional response model? [duplicate]

Can anyone tell me the theory behind fractional response model, how it really works? I wonder if the logistic regression works only with binary variable {0,1}, why when conducting a GLM with ...
1
vote
1answer
56 views

Strange output for pairwise comparisons on glm with quasi-binomial distribution

I'm new to CrossValidated - I've read up on how to ask questions properly but sorry if I do anything slightly wrong. My data is showing whether microplastics were present or absent in the gut of fish ...
1
vote
1answer
166 views

“scale” in logistic regression

I am working on translating some R code into Python's statsmodels package, chiefly some logistic regression work that I've done, when I came across the following in ...
0
votes
0answers
74 views

Two intercepts for zero-truncated negative binomial model using VGAM

I am trying to understand the first and second intercept for the zero-truncated negative binomial regression model I estimated using VGAM. Below is my syntax: mod.negb <- vglm(ED_Visit ~ Male + ...
0
votes
0answers
17 views

How can I compare coefficient values from two different count models based on the same variables?

I'm running two quasi-Poisson models with the exact same variables, but on samples from two different countries. I'm doing this because I'm particularly interested in seeing whether the relationship ...
2
votes
1answer
119 views

Why ever use a quasipoisson model instead of bootstrapped poisson GLM?

A poisson GLM and a quasipoisson regression model will given identical point estimates for the beta parameter of the linear predictor. The quasipoisson model is typically used when there is ...
1
vote
0answers
414 views

Quasi-binomial GLM in R

I'd like some advice on data I'm analyzing from a factorial-design study in which each sample is a count of 200 urchin eggs that were exposed to various types and concentrations of pollutants, and for ...
1
vote
0answers
51 views

How to Calculate Sample Size Requirements for Quasi-Poisson / Negative Binomial AB Test in R

I'm doing an AB test of two conditions, where the success metric is looking for an increase in the number of users taking a particular action each day. The variance is hugely larger than the mean, ...
4
votes
1answer
340 views

Does poisson penalized quasi likelihood regression use biased estimators?

A professor told me the Poisson glmmPQL (mixed-effects/hierarchical) regression gives biased estimates. The paper https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3886992/ from PLOS One says that ...
2
votes
1answer
227 views

What is the distribution in Quasi-Poisson regression?

For Poisson regression, the assumption is that Y has a Poisson distribution. Is the same assumption true for Quasi-Poisson regression?
2
votes
2answers
1k views

GLM model interpretation of quasi poisson incorporating dispersion in R

I am fairly new to GLMs, and am currently working with a quasipoisson distribution due to overdispersion. I have read through the relevant sections of Mixed Effects Models and Extensions in Ecology ...
1
vote
1answer
33 views

What is the impact of excess zeros on poisson regression coefficient estimates?

The background I have a dataset with some zeros - based on how I segment my data, it is either 50% of the observations or 80% of the observations. The data is not actually count data, but from what i ...
0
votes
1answer
342 views

How to fit a zero-inflated quasi-poisson distribution to a continuous variable using GAM [closed]

I am trying to fit a GAM to a continuous variable which is zero-inflated. However, since my variable is continuous, I am not able to use ziP() for a zero-inflated quasi-Poisson. Is there someway to ...
1
vote
1answer
789 views

Quasi-Poisson - Goodness of fit

Can I use likelihood ratio chi-squared tests to test the goodness of fit for quasi-Poisson models?
3
votes
1answer
193 views

Why the absence of probability distribution for using Quasi-likelihood?

We can compare $2\cdot(\text{difference in log-likelihood})$ to a chi-square distribution but why can't we find or invent a new distribution whose test statistics are some function or form involving ...
1
vote
0answers
253 views

Interpretation of fractional regression (GLM quasibinomial with logit link) coefficients

I am writing a research paper commenting theresults of the following regression, which is a GLM quasibionomial regression with a logit link (the outcome variable capfactor ranges between 0 and 1). <...
1
vote
1answer
87 views

Quasi/Complete separation due to huge and infinite values

(R statistics) My question is regarding this warning. My data contains patients and healthy subjects. Exponential decay is my outcome measure. I have a example dataset here I managed to run ...
1
vote
0answers
126 views

Is my overdispersion too large in this quasibinomial model?

I have used a quasibinomial model on my data, but my overdispersion coefficient seems to be too large with a value of 40.78776. ...