Questions tagged [zero-inflation]

Excessive 0's in a variable compared to a specified reference distribution. Regression approaches include zero-inflated models and hurdle (2-part) models. For count data, zero-inflated and hurdle models based on Poisson or negative binomial distributions are common (ZIP/ZINB and HP/HNB).

Filter by
Sorted by
Tagged with
0 votes
0 answers
15 views

Understanding the Binomial Component of the Zero-Inflation Model and Why zeroinfl in R Returns NA for features in the Summary

I am trying to apply a Zero-Inflated Negative Binomial model to a dataset in R using the zeroinfl() function from the pscl ...
user avatar
  • 227
0 votes
0 answers
23 views

DHARMa problems: Poisson or Zero-inflated negative binomial model?

I have a dataset of abundance of of rodent (AA) in 63 sites, which located in two main area (northern and southern part, NS). AA ...
user avatar
0 votes
0 answers
30 views

Bayesian Additive Regression Trees: Zero-inflated explanatory variable, will it influence the model and variable selection?

I am currently implementing BART to model the distribution of a marine species (using the embarcadero package). I am using environmental covariates, but also some prey data that are very-much zero-...
user avatar
  • 321
1 vote
0 answers
18 views

Methods to handle an arbitrary peak within a distribution

I had someone come to me for advice on how to deal with their weird data. The predictor variable is categorical (experimental condition), and the outcome variable is continuous (amount of money in a ...
user avatar
1 vote
2 answers
46 views

How to multiply two models together in R?

I am constructing a two-stage GAM to model capture rate across my study site while accounting for zero inflation. ...
user avatar
  • 11
0 votes
0 answers
7 views

Options for two-stage GAM univariate model selection when AICc values <2Δ

I am constructing a two-stage GAM (stage 1 presence/absence with binomial and logit link, stage 2 abundance with poisson and log link) to model capture rate across my study site. As this is a pilot ...
user avatar
  • 11
1 vote
0 answers
10 views

appropriate method for over dispersion of zeros, but not count data (0/1 did you move; 0/1 did you move to x town)

I am looking at a migration outcome question. Basically, it's did you move (0/1); and if you did move, did you move to x town (0/1), as a function of tenure, economic conditions, etc.. I am somewhat ...
user avatar
  • 155
0 votes
0 answers
12 views

What statistical test is appropriate when the independent variables are binary but tied to multiple sites and the dependent variables has many 0s?

I have a data set of microplastic particles/kg taken from various beaches. Each beach has several binary criteria which are the independent variables of the study. My issue is that in some samples ...
user avatar
1 vote
1 answer
33 views

Using binary dummy to control excess of zeros in continuous data

I'm running an OLS and one of the explanatory variables accounts the sewage coverage, which is zero in 42% of the observations (these are true zeros). I'm worried that the linear effect is not going ...
user avatar
  • 23
0 votes
1 answer
39 views

Analysing a small data set with lots of zeros? Dependent variable is (semi?) continuous, one fixed (categorical) factor

Hoping to get some advice for choosing an appropriate model for my data. I have a small data set looking at the level of a particular protein (heat shock protein 70) in individuals exposed to ...
user avatar
2 votes
1 answer
32 views

Regression in data with one group, having just zeros as outcome

I have a data set, consisting of positive and negative patients (virus infection). If the patient is negative, it has 0 as outcome (y), if it is positive it has a positive value, up to 100. The input (...
user avatar
  • 21
1 vote
0 answers
35 views

How do I determine which count model (quasipoisson, negative binomial, or zero-inflated GLM in R) best fits my count data?

Background: I am modeling counts per square meter of an invasive species, which range from 0 to 380, as part of my master's thesis. The data set has 50 observed zeros out of 161 observations. After ...
user avatar
3 votes
2 answers
38 views

Generate values from a zero inflated negative binomial fit

I am using the pscl package to fit a simple zeroinf(...,dist = "negbin"). I would like to simulate the probability ...
user avatar
  • 943
0 votes
0 answers
9 views

which type of model to predict percentages with an almost bi modal distribution

I wonder, which type of regression (?) model I could use for a dependent variable corresponding to percentages with an almost bi modal distribution like this? The percentages correspond to weekly room ...
user avatar
  • 1,306
0 votes
0 answers
31 views

Small N analysis for a field experiment question, and rare events logit

so, I am running a "Field" experiment, meaning, people in a real life situation are subject to certain treatment, and I am trying to measure the effect of said treatment. In this case its an ...
user avatar
  • 3
0 votes
0 answers
35 views

Cross validation and aggregated data for count models

I know that this question has already been formulated in other posts where there are quite comprehensive answers, I refer especially to these two explanations (thread1, thread2). However I continue to ...
user avatar
0 votes
0 answers
13 views

In a zero-inflated Poisson model, what is the relationship between the rate, the 0-probability, and dispersion?

Suppose a random variable $Y$ is distributed as a two-parameter zero-inflated Poisson. That is, $Y=XW$ with $X \sim \text{Poisson}(\theta)$ and $W \sim \text{Bernoulli}(\nu)$. Suppose I obtain a ...
user avatar
  • 53k
1 vote
1 answer
66 views

Zero inflation with no zeros in data?

I am investigating feeding rates in great apes. Because I am using count data, I built a generalized linear model with a Poisson link function. So I put the number of items in a bundle as a response ...
user avatar
0 votes
0 answers
18 views

Should heteroscedasticity of "residuals" frighten?

I have a dataset with 10 variables. A couple of them are "Customers" and "Sales". I want to find what is the increase in sales with a unit increase in customers. That is, I want to ...
user avatar
1 vote
1 answer
50 views

How to deal with zero-inflated data with negative values

Participants in a survey answer two questions similar to this: What is your estimate, how on average a woman performs in this test (from 0 to 100)? What is your estimate, how on average a man ...
user avatar
1 vote
0 answers
29 views

Help with interpretation of a zero-inflated gamma model

Background: I'm in fairly over my head here, working on a project for class where we had to choose our own publicly available dataset to analyse and I chose one with a zero-inflated continuous ...
user avatar
  • 11
0 votes
0 answers
21 views

Zeroinflated and degree of freedom

I am using zeroinfl test with my data, and was wondering if it is enough to mention the z value in a paper. Otherwise, I used the lrtest to calculate the X² and p values as follows: model1<-...
user avatar
  • 1
0 votes
2 answers
64 views

Can anybody think of a transformation to normalize these data?

I want to do a multiple linear regression but I do not find any way to normalize the data I have. This is the distribution without transformation. The data contains a good number of zeros. When I ...
user avatar
0 votes
0 answers
12 views

Another hurdle models question

Sources of zeros are usually identified as “random zeros” and “structural zeros.” Random zeros may occur from sampling variability. Structural zeros may occur due to sampling at a place where the ...
user avatar
  • 11
0 votes
1 answer
56 views

How to improve the fit of a beta zero-inflated regression model (GAMLSS)?

I'm working with a response variable with values between 0.0 and 1.0. I have a lot of zero. Thus, I'm using beta zero-inflated regression model. Specifically, I'm using the function gamlss from the ...
user avatar
0 votes
0 answers
51 views

manually calculated the fitted values of gamma hurdle model from glmmTMB

I'm manually calculating the fitted values from a gamma hurdle model and compare the results with glmmTMB package. I transferred the response variable to be zero and non-zero and fitted the logistic ...
user avatar
0 votes
0 answers
18 views

Best way to code ratio of two skewed, heavy-tailed continuous variables with zero values present in denominator?

I'm modeling the ratio of total medical cost in one 12 month period and the subsequent 12 month period (total_medical_cost_ratio = post_total_medical_cost/pre_total_medical_cost). The data is skewed, ...
user avatar
  • 4,504
1 vote
0 answers
23 views

Inclusion of values that can only be zero in Poisson glmer

I have carried out a biological experiment in which the dependent variable is female propensity to mating, measured as the number of times a female animal accepts a mating attempt from a male. I have ...
user avatar
  • 21
1 vote
1 answer
38 views

Modelling the difference of two zero-inflated count variables

I'm trying to model count data on house sparrows numbers before and after mitigation measures and want to find out which of these measures affect their numbers (if any). I'm however not entirely sure ...
user avatar
  • 11
1 vote
1 answer
76 views

Interpret interaction coefficients in a negative binomial

Trying to understand count distributions of a fish species over two habitat types (class variable) and three seasons (class variable). The table below shows the results of a hurdle negative binomial ...
user avatar
  • 11
0 votes
0 answers
17 views

Can a variable derived using count data be used to model zero inflation in a Zero-Inflation Poisson model?

Say I have a zero-inflation Poisson model that is count ~ var1. My count data is inflated by zeros because count is the number of symptoms of a particular disorder. If count > 5, and those symptoms ...
user avatar
1 vote
0 answers
36 views

A mixture of discrete and continuous components

Suppose that random variable $X$ is sampled from Bernoulli with probability $\pi$. Let $Y\sim N(\mu, \sigma^2)$ and denote $Z=XY$. Then, when $X=0$, $Z$ equals 0 (and this happens with probability $1-\...
user avatar
  • 789
1 vote
2 answers
96 views

How to model a probability distribution to phone call duration?

I was plotting the phone call durations, and the result seems like a right-skewed histogram. Shape-wise, Poisson, Raylegh, Chi-2 can be candidates for modeling that. But is there a more rigorous way ...
user avatar
  • 141
0 votes
0 answers
16 views

Mixed model with plenty of 0's and a unique fixed effect per a random factor

I am having trouble analysing some of my data. I'm trying to test the effect of treatment (Type) with three levels (Control/Threarment1/Threatment2) on the abundance of Arthropods belonging to ...
user avatar
1 vote
1 answer
69 views

Transform variables with zero-inflated values and positive skewness

I have over 30 features: several have zero-inflated and highly positive skewed distribution. Those distributions are expected because they are semi-continuous monetary related features. For example: ...
user avatar
0 votes
0 answers
11 views

Comparing model types: zinb and nb

I am modelling variation in the count data which represents number of boats in images. There are a lot of images with 0 counts, and the variance is greater than the mean, so after comparing poisson, ...
user avatar
1 vote
0 answers
19 views

GLM with zero inflated negative binomial distribution is underdispersed

I am trying to model count data which is the number of boats present in images. The variance in this data is greater than the mean, and the data is zero-inflated. I compared poisson, negative binomial,...
user avatar
0 votes
0 answers
15 views

How to perform model diagnostics on ZINB?

I have the following model: ...
user avatar
6 votes
1 answer
544 views

Zero-inflated Gaussian for weights below zero recorded as 0?

I'm aware of the general idea behind zero-inflated model, and have used the zero-inflated Poisson and negative binomial. However, the data I currently have has a little different format that makes me ...
user avatar
1 vote
1 answer
86 views

How does PSCL fit zero-inflated negative binomial models?

I've been using the PSCL zero-inflated negative binomial model to simply fit count data. I'm not specifying any complex formula, essentially I just provide the count data and try to find the ...
user avatar
  • 11
1 vote
2 answers
110 views

Mean difference analysis for a paired data with zero inflation. Can Wilcoxon test work?

I have a set of data that represent the sum of white matter streamlines termination in 64 regions of interest in the brain surface collected from 40 subjects. Below is an example of a paired data of ...
user avatar
3 votes
1 answer
163 views

zeroinfl doesn't seem to work (R)?

I´vd tried to fit a zero-inflated negative binomial model with zeroinfl (package pscl): ...
user avatar
0 votes
0 answers
54 views

Proper regression for my dataset - Zero-inflated beta regression?

I'm a bit stumped on the regression that I should be using for my dataset. There seems to be a general trend that as distance to a location decreases, the percentage of red material increases (as a ...
user avatar
2 votes
1 answer
33 views

Converting zero count regression model back to frequency data?

I’m new to statistics but I’m performing a systematic review into exercise and alcohol intake. One of the papers uses a zero inflated count regression model to report the alcohol consumption per day. ...
user avatar
1 vote
1 answer
141 views

DHARMa testZeroInflation: how to interpret output?

I'm analyzing count data with a negative binomial GLMM via the R package glmmTMB and lme4. I'm running DHARMa diagnostics, one of which is testing for zero inflation and I'm having some trouble ...
user avatar
  • 27
0 votes
0 answers
16 views

How to deal with dispersion and Zeroninflation in GLMM binomial models?

I am working with a GLMM (binomial family). I test my model for dispersion and ZeroInflation with the DHARMa package. Here is my code to show what I test : ...
user avatar
  • 1
0 votes
0 answers
53 views

Hurdle model: How to compare slopes?

I'm analyzing the change in seagrass cover (count data of shoots) between two years of different species. Therefore, I made a model similar to this: ...
user avatar
  • 43
1 vote
1 answer
224 views

Interpretation of Positive Count Coefficients in Hurdle Model

What is the proper interpretation of the coefficients for the positive count part of a hurdle model (truncated Poisson or Negative Binomial)? I have read that the interpretation of the coefficients ...
user avatar
  • 45
1 vote
0 answers
20 views

Is Hurdle model recommended if there is a few zero counts?

Is Hurdle model recommended if there is small number of zero counts? Is there any limit for the zero count to fit Hurdle model?
user avatar
1 vote
1 answer
68 views

ZINB or ZIP for cetacean data with many zeros [duplicate]

I am stuck in my thesis to conduct a GLM with cetacean data (count, density, effort and covariates). I have over 94% zeros in my dataset meaning I should be using a GLM with zero-inflated negative ...
user avatar

1
2 3 4 5
11