Questions tagged [negative-binomial-distribution]
A discrete, univariate distribution modelling the number of ${\rm Bernoulli}(p)$ trial successes until a specified number of failures occur.
989 questions
6
votes
1
answer
239
views
Choosing an appropriate mixed model for zero-inflated, right-skewed PTSD symptom score (Tweedie vs Negative Binomial vs Gaussian vs ?)
Question: I am analysing longitudinal survey data measuring PTSD symptoms (PTSS) among police officers using R, and I am uncertain which mixed modeling framework is most appropriate for the outcome. ...
- 61
7
votes
2
answers
206
views
Determine if a negative binomial glm shows zero inflation from diagnostic plots
I have found many similar questions, but my diagnostic plot results seem to show problems that are too specific for me to interpret.
I have the following data :
...
- 73
1
vote
0
answers
33
views
Is g-computation using mixed-effects models with time varying confounders causally valid?
I am trying to clarify the conditions under which g-computation based solely on an outcome regression is a valid causal estimator in longitudinal data; specifically, a mixed-effects (or more generally,...
- 350
4
votes
1
answer
173
views
Choosing count model with marked overdispersion and zero-inflation
Background
In a retrospective cohort study, I aim to identify covariates associated with overlapping prescriptions of a given drug (a proxy for misuse). This is a descriptive study (I do not perform ...
- 600
5
votes
2
answers
179
views
Over-dispersion parameter for a negative binomial distribution
A paper states that the annualized bleeding event rate is 39.86 with a 95% CI of (33.05,48.07). This is estimated based on a sample size of 37 and with the exposure time as 0.5 years for each patient. ...
- 125
1
vote
0
answers
19
views
How to interpret DHARMa residual diagnostic showing bimodal pattern (“two humps”) in a negative binomial GLM?
I’m trying to fit a negative binomial GLM for a count response variable (stems per hectare). The data were significantly overdispersed. Thus I chose glm.nb() from MASS and then checked model fit ...
- 11
1
vote
0
answers
53
views
Repeatability (ICC) for negative binomial and beta models
I am estimating repeatability (a type of intra-class correlation) from mixed models with negative binomial and beta distributions and want to confirm that the variance transformations to obtain ...
- 11
0
votes
0
answers
51
views
glmmTMB warns of false convergence with betabinomial models but no with bin or negbin models
I've been looking for the best distribution to fit a glmm model to my data. The best seems to be a betabinomial model, but it gives a warning about false convergence, which is caused by large z values:...
- 163
7
votes
1
answer
294
views
How to deal with missing data in the response variable
I am writing to seek your expertise with a project I am working on, which is described briefly here:
The data are on disease occurrence ('number of cases' is my dependent variable) from 1960 to 2025. ...
3
votes
1
answer
122
views
Propagating uncertainty while calculating cumulative timeseries data (negative binomial error distribution)
I am trying to appropriately propagate uncertainty from my predicted datasets following a GAM model. Essentially, I have weekly counts of data (i.e., egg counts), but want to calculate cumulative ...
- 33
3
votes
0
answers
93
views
No indication data is overdispersed but model fit improves when accounting for it
In glmmTMB I've noticed on several occasions that I can have a standard Poisson glmer with no indication of overdispersion using the check_overdispersion, ...
0
votes
0
answers
62
views
Permutation-Based Multiple Testing Correction in Negative Binomial Regression
I am conducting a simulation-based analysis to test potential associations between genetic variables and a dependent outcome.
My model meets the assumptions for a negative binomial regression.
I would ...
1
vote
1
answer
97
views
Huge coefficient in negative binomial regression model
I'm estimating a negative binomial regression model with MASS::glm.nb
A coefficient for one level of a factor variable is very huge.
It equals 30, which after being ...
- 3,488
6
votes
1
answer
156
views
Implementations of profile likelihood intervals for negative binomial regression
I am trying to understand the limitations of profile likelihood intervals in small datasets. I was trying to do a small simulation study for negative binomial regression but it turns out common R ...
- 5,403
9
votes
2
answers
469
views
Why/when will negative binomial regression flip the sign of the effect from analogous Poisson regression, after covariate adjustment
First, to be clear, the issue mentioned in the title does not happen often and, in every parametric simulation I've tried, both NB and Poisson produce similar estimates, regardless of the type of ...
- 91