Questions tagged [gee]

Stands for Generalized Estimating Equations which is an approach to estimating regression coefficients. GEE can be used on clustered / longitudinal data and has the attractive property that it provides consistent estimators of regression coefficients and unbiased inference even when the association structure within a cluster is mis-specified.

Filter by
Sorted by
Tagged with
6
votes
1answer
111 views

Why not always use generalized estimating equations (GEE) instead of linear mixed models?

I read about generalized estimating equations (GEE) here, here and at other sites. It is mentioned in first of above links that "the parameter estimates are nearly identical" for linear ...
3
votes
1answer
42 views

Alternative to Mixed ANOVA without homogeneity of variances

As is tradition on these posts, I should say I'm relatively new to statistical analysis at this level so if I don't provide enough info off the bat bear with me. So I've conducted an experiment ...
0
votes
0answers
4 views

Analysing change in preference (proportion time, 3 repetitions)

I thought I'd ask for help here as I'm getting myself confused! In a nutshell, I am looking at the proportion of time animals spend in 4 sound conditions (each animal is tested 3 times) as a measure ...
0
votes
1answer
25 views

May Pattern-Mixture Models be used in conjunction with generalized estimating equations in addition to with mixed-models?

May Pattern-Mixture Models be used with generalized estimating equations for longitudinal data analysis? Hedeker et al mentions pattern-mixture models for handling non-ignorable drop-out patterns and ...
1
vote
0answers
8 views

May the fisher-scoring algorithm be used when we can only compute the quasi-likelihood?

May the fisher-scoring algorithm be used when we can only compute the quasi-likelihood and can't find the full-likelihood function? IE, some cases where it's advantageous to use GEE rather than Mixed-...
0
votes
0answers
13 views

Confusion about marginal vs conditional models and reversal of association

I've been reading Walter Stroup's Generalized Linear Mixed Models: Modern Concepts, Methods, and Applications (2013), wherein the difference between "marginal" vs "conditional" models and inference is ...
1
vote
1answer
38 views

Post hoc test after GEE with repeated measures using R

I'm working on repeated measures design GEE and would like to get the result of post-Hoc test, but I don`t know how to do it. I just used those methods to get the result of post-Hoc test, but it ...
2
votes
1answer
69 views

How to deal with skewed distribution in GEE or GLM?

I have a group of repeated measurement data, the dependent variable (y) was skewed distribution when I build the GEE, should I transform y to a normal distribution variable first? Or Can I build the ...
0
votes
0answers
6 views

Are generalized estimating equations appropriate to use in this situation?

I’m an undergraduate doing research under a grad student. I’m trying to see if we’ve used generalized estimating equations correctly and if it’s appropriate for what we’re trying to figure out. ...
0
votes
0answers
8 views

GEEs for temporal autocorrelation and the ACF plot

I have a dataset with temporal autocorrelation in it. The response variables consists of 0s and 1s, and I used a binomial GLM first. However, when I used: residualsmodel <- resid(model) acf(...
0
votes
0answers
14 views

Panel Data Regression

Here (attached image), FTAs is my "Response Variable" and all the columns in the right are my "Covariates". It is clearly dependent on time so, there's a strong correlation. I want to find the most ...
0
votes
0answers
22 views

Comparing highly correlated predictors in log-binomial regression

I'm trying to figure out the best way to assess which of my highly correlated independent variables best predicts my dependent variable (y), a binary variable coded ...
0
votes
0answers
10 views

GEEGLM binomial modelling with Probit or logit

I am fitting a GEE model in R (multivariate binomial) There are no repeated measures and assume there is a within cluster correlation. If I use a probit link, do I have to exponentiate the output ...
0
votes
0answers
40 views

In R, does passing the IPW weights to the geeglm function in the geepack manage the Missing At Random (MAR) scenario?

Dear R users good at statistical modelling, please help me with this question. I want to use GEE in my analysis. Unfortunately, I have missing observations. It seems it is at least MAR, as I cannot ...
0
votes
0answers
72 views

If paired t-test tests mean differences rather than difference in means, what does GEE or mixed models followed by ANOVA analyze?

Paired t-test takes differences between data in two groups, say, time points. A subject must have both information: "pre" and "post". Otherwise the difference cannot be computed. So, the mean change ...
1
vote
0answers
20 views

GEE Std. Error Estimators: Pros and Cons of using jack-knife or bootstrap estimator of std. errors rather than sandwich when clusters$>30$?

What are the Advantages and disadvantages of using jack-knife, bootstrap estimator rather than sandwich (Huber-White) estimator in the context of generalized estimating equations? I heard sandwich ...
0
votes
1answer
51 views

GEE: results significant only with independent and exchangeable correlation structure

I am trying to fit a GEE Poisson model on a panel dataset consisting of T=360 and N=304 for a total of >108,000 observations in Stata. My response variable measures a count of people imprisoned, and I ...
0
votes
1answer
21 views

Deriving asymptotic variance of generalized estimating equation estimator (GEE)

As well known to us, K.Y. Liang and S. Zeger proposed GEE for longitudinal data analysis in their famous paper[1]. At the appendix of the paper, authors show the proof of Theorem 2. I tried to ...
0
votes
0answers
54 views

What exactly does it mean “conditional to random effect”?

I understand, that in (generalized) mixed models, the calculated $\beta$s are not population ones, but rather "conditional to the random effect". Let's say, I have a model, where I assess a set of ...
3
votes
1answer
97 views

What does it mean exactly to have marginal (population-averaged) or subject-specific effects, say in mixed models vs. GEE?

I know that mixed-effect models produce subject-specific answers and the GEE produces marginal effect. But I don't get it very well still, I'm sorry for a silly question. Let me explain what exactly ...
2
votes
2answers
50 views

Is it possible to use location-scale family distributions for mixed effects modeling?

Is it possible to use location-scale family distributions for mixed modeling or generalized estimation equations? The only location-scale family package I know of is GAMLSS in R which is for additive ...
1
vote
2answers
39 views

Is it valid to use a difference score between sequential timepoints as an independent variable in a longitudinal regression analysis?

Is it valid to use change scores, i.e. like change in body weight taken between consecutive time points in longitudinal analysis where I'm using nlme/gee? For example, bodyweight at hour 3 minus ...
0
votes
0answers
20 views

How to compute the prediction interval with poisson GEE?

How do I compute using R the prediction interval for poisson GEE via geeglm? I prefer geeglm because of the "waves" argument. Aside: I'm trying to predict what future counts of an event will be in ...
0
votes
0answers
17 views

Is it possible to do cross-validation with GEE?

Is it possible to do cross-validation with GEE and recursive partitioning trees? I wanted to use recursive partitioning on longitudinal data and them run the data through gee.
2
votes
1answer
104 views

May “offsets” be used in mixed-effects poisson regression?

Mixed-effects poisson regression studies counts for example of the incidence of a disease given the individual's random-intercept/slope. Mixed-effects regression studies individuals rather than ...
1
vote
0answers
21 views

Does poisson GEE require “scale.fix=TRUE” and “scale.value=1”?

Does poisson gee require scale.fix=TRUE and scale.value=1 for the package geepack? Aside: I heard gee package in r doesn't allow one to specify the scale.value and only let's one specify scale.fix=T/...
0
votes
1answer
42 views

Model validation with GEE-GLM

I am trying to find some resources on model validation for GEE-GLMs. Unfortunately I can't afford to purchase expensive textbooks and many of the books address GEEs using other software such as Stata. ...
0
votes
0answers
24 views

GEEs and GLMs: what can I do about the auto-correlation?

I want to run a model that somehow incorporates temporal autocorrelation. I have a dataset which consists of the minutes of Species A vocalizing at a single site. This is data recorded continuously ...
1
vote
1answer
58 views

Interpretation of mixed-effects versus GEE poisson/logit regression parameters

What's the difference in the interpretation of the model parameters (intercept +slope) in the mixed-effects model and GEE model for poisson and logistic regression?
0
votes
0answers
47 views

What do the different correlation structures mean with GEE?

What are the differences between the common options for correlation structures are and when do we use them, e.g. "independent", "exchangeable", "autoregressive", "unstructured", "fixed", "stationary", ...
1
vote
0answers
141 views

Is this possible to fit a MMRM (SAS REPEATED) with compound symmetry or AR1 model with glmmTMB?

MMRM (Mixed effect Model Repeat Measurement) models are special cases of the mixed models, where no random effects are used, only the residual covariance is modelled. This is commonly used for ...
0
votes
0answers
16 views

Multilevel versus clusteredness in the sample

does anyone can explain to me the main differences between multilevel estimations and clustering correction? I think with HLM technique we can get estimation to the random parameters (the group level ...
0
votes
1answer
66 views

Inclusion of clustering within a Cox PH model

I am working with multi-center data and modelling time-to-event data. I have included a cluster term within my model to account for similarities between patients that were treated at the same center. ...
1
vote
1answer
81 views

Advantages of using Cox model versus logistic GEE?

What are the advantages of using Cox proportional hazards model versus logistic GEE? Can one generate Kaplan-Meier remainder estimator curves using logistic GEE?
2
votes
1answer
99 views

How can the clustered robust standard errors be smaller than the model based ones?

I ran a GEE model and I used it to check the difference between the empirical standard errors and the model-based one. For almost the variables, the empirical standard error was greater than the OLS ...
1
vote
1answer
131 views

GEE estimates in R diverge when all responses of one level of a categorical predictor are 0

This is my first post so please bear with me. I do not have statistics background and I am still learning my way around so your answers will be very helpful. I am using GEE in R to compare the levels ...
1
vote
1answer
112 views

Continuous outcome variable - Select between GEE and Linear Mixed Models

What are the arguments for and against selecting GEE and Linear Mixed Models when the outcome variable is continuous? Are they any circumstances where one performs better than other? The data I am ...
2
votes
0answers
30 views

Is GEE a good replacement for ANCOVA to analyze change from baseline on 8-points Likert item?

I was asked to analyze adjusted change from baseline. The variable of interest is a customer answer reported on 8-point Likert scale, but constructed from numbers 0 - 8, rather than labels ("weak", "...
0
votes
0answers
11 views

geeglm estimates interpretation

I would like to understand how to interpret the geeglm estimates for a gaussian family model. Say the formula is Dependent ~ A Dependent ~ B And the results shows [1,] "A" "-3.60767907868019e-...
0
votes
0answers
101 views

Post hoc test after a generalized estimating equation (GEE)

It would be great if someone could provide any clue about performing the post hoc test after a significant gee model. In my own case: We have deployed temperature sensors below the canopies of 3 ...
0
votes
0answers
20 views

Should I use a GEE model for a mixed logistic regression problem with a multinomial outcome?

I'm trying to model linguistic data where my hypothesis assumes that a multinomial categorical response from a subject is a function of fixed effects (gender and subject's occupation, say) and, since ...
0
votes
0answers
11 views

comparing coefficient in increamental models with changing sample sizes

I have a panel data with 50 features over the course of 40 years. I want to evaluate the success of musicians. These 50 features cover 4 aspects of individuals careers. To uncover the importance of ...
2
votes
1answer
88 views

How is nlme package in R “non-linear”? [closed]

Why is the package nlme non-linear? Is it non-linear in the sense that gee is non-linear because of correlation structure/covariance pattern like exchangeable/AR1? So if we drew out correlated errors ...
1
vote
0answers
80 views

To GEE or not to GEE

I am post doctoral fellow in neurosciences. I have a quite basic question that I failed to solve by myself. In a pretest experiment for an fMRI study, we have asked to 3 age groups (~30 ...
3
votes
2answers
162 views

GEE logit / Poisson versus mixed effects Poisson / logit

There's a way to do Poisson or logit mixed effects and Poisson or logit GEE in R. What's the difference between GEE and the mixed effects models for Poisson / logistic regression? I heard its the ...
0
votes
0answers
27 views

How to build a model in GEE using poisson?

I am going to look for the association of summer temperature with different health outcomes. I have to use Poisson regression using GEE. These are time-series data and all the predictor variables ...
0
votes
0answers
49 views

How could I get Prediction Interval from GEE model? Or how could I constructed using formulas?

I'm currently using GEE with Gamma link to fit model. The goal is to predict response variable for new data, I tried "Predict" function in r, but I also need Prediction Interval. I appreciate all the ...
1
vote
2answers
71 views

Why estimate using GEEs inspite of their disadvantages over using ML?

I know sometimes I want to know the population-level estimates, but the problem with GEEs is I can't calculate the likelihood, and therefore all models I make with it aren't comparable, and I don't ...
2
votes
1answer
123 views

Seeking to understand using the Firth correction in Generalized Estimating Equations to deal with quasi-complete separation

In order to deal with complete separation in my data someone suggested that I run penalized GEE (PGEE) by adding a Firth-type penalty term in R. Although I have read many papers on the Firth ...
0
votes
1answer
51 views

Correcting for bias in GEE models with small cluster size

In GEE, several methods have been proposed for correcting for bias when the cluster size is small to moderate (<40). Some have proposed alternative variance estimators, e.g. Morel, Bokossa, and ...

1
2 3 4 5 6