#GLMM
LightNews LightNews
lololosparkk.bsky.social
You just know damn well it's going to end up looking like some average quirky and egdy GLMM bsky.app/profile/puny...
October 29, 2026 at 6:29 AM
palmtree-sloth.bsky.social
He can be both it’s okay… they can be the secret hybrid royalty glmm… I always saw him as an angel sense he acts as a priest and I assume pride at wanna go to heaven … BUT a demon also works cuz he worked in the media industry and they’re all going to hell
October 16, 2026 at 12:50 AM
paperposterbot.bsky.social
arXiv📈🤖
Deep Neural Mixed Models for Nonlinear Longitudinal Prediction with Subject-Specific Heterogeneity
By Hasan, Ramezani, Chakraborty
Longitudinal health data often exhibit within-subject correlation, subject-specific heterogeneity, and complex nonlinear predictor--response relationships that may not be adequately captured by conventional mixed models. We propose a Deep Generalized Neural Network Mixed Model (DGNMM) that extends the GLMM framework by using a deep neural network to learn nonlinear population-level relationships while incorporating directly trainable, regularized subject-specific effects to capture individual heterogeneity. The model is estimated using prediction-oriented stochastic optimization with separate regularization of the neural-network parameters and subject-specific effects. This formulation allows us to investigate when increased nonlinear representation capacity provides additional predictive benefit in the presence of subject-specific heterogeneity. A simulation study varies population-level nonlinearity and subject-specific heterogeneity and compares DGNMM with GLMM, artificial neural network (ANN), Maity--Pal (MP), and GNMM alternatives for continuous and binary longitudinal outcomes. In the real-data application, DGNMM achieved the lowest MSPE for continuous depression and anxiety prediction, while alternative models remained competitive for binary outcomes. Exploratory subject-level analyses showed heterogeneous predictive gains across individuals, but the magnitude of the estimated subject-specific effects was not significantly associated with predictive gain. Overall, increased nonlinear representation capacity was most beneficial for continuous prediction under nonlinear response structures, while greater subject-specific heterogeneity did not systematically yield larger predictive gains.
October 15, 2026 at 1:54 AM
tedpavlic.bsky.social
@provete.bsky.social -- Thanks, again, for the link to that nice resource! Motivated it, I've added a few more entries under the "Classic tests as GLMM" section in the GLMM Analyzer page (linked again below), and I've added the site itself under the "References"
tpavlic.github.io/topic_visual...
GLMM Analyzer – generalized linear mixed models in the browser
Fit generalized linear models, and mixed models with one grouping factor, to a table in the browser: normal, binomial, Poisson, negative binomial, gamma, beta, and ordinal responses with interactions ...
tpavlic.github.io
October 13, 2026 at 12:59 AM
paperposterbot.bsky.social
arXiv📈🤖
Deep Neural Mixed Models for Nonlinear Longitudinal Prediction with Subject-Specific Heterogeneity
By Hasan, Ramezani, Chakraborty
Longitudinal health data often exhibit within-subject correlation, subject-specific heterogeneity, and complex nonlinear predictor--response relationships that may not be adequately captured by conventional mixed models. We propose a Deep Generalized Neural Network Mixed Model (DGNMM) that extends the GLMM framework by using a deep neural network to learn nonlinear population-level relationships while incorporating directly trainable, regularized subject-specific effects to capture individual heterogeneity. The model is estimated using prediction-oriented stochastic optimization with separate regularization of the neural-network parameters and subject-specific effects. This formulation allows us to investigate when increased nonlinear representation capacity provides additional predictive benefit in the presence of subject-specific heterogeneity. A simulation study varies population-level nonlinearity and subject-specific heterogeneity and compares DGNMM with GLMM, artificial neural network (ANN), Maity--Pal (MP), and GNMM alternatives for continuous and binary longitudinal outcomes. In the real-data application, DGNMM achieved the lowest MSPE for continuous depression and anxiety prediction, while alternative models remained competitive for binary outcomes. Exploratory subject-level analyses showed heterogeneous predictive gains across individuals, but the magnitude of the estimated subject-specific effects was not significantly associated with predictive gain. Overall, increased nonlinear representation capacity was most beneficial for continuous prediction under nonlinear response structures, while greater subject-specific heterogeneity did not systematically yield larger predictive gains.
October 14, 2026 at 9:17 PM
gsimpson.bsky.social
If you fit models with glmmTMB(), gratia 📦 has gained support for these model objects, including glmmTMB's nascent capability for fitting penalized spline terms using mgcv's s()

You'll need the GitHub versions of both gratia & glmmTMB to try this out #RStats #statistics

github.com/glmmTMB/glmm...
Four panels from gratia’s draw() method for a glmmTMB model: a curved smooth effect of x with a shaded confidence band, separate Gaussian QQ plots for site random intercepts and random slopes of z, and the linear parametric effect of z with a shaded confidence band.
October 12, 2026 at 5:42 PM
bbolker.bsky.social
Thanks! This inches us toward `glmmTMB` being a replacement for `gamm4` - which, with respect to its authors, I've found a headache to work with (huge model objects, lack of downstream methods, etc.). BUT `glmmTMB` still has issues with predictions, starting values, etc. github.com/glmmTMB/glmm...
open thread for mgcv smooth issues · Issue #928 · glmmTMB/glmmTMB
See notes. Currently known outstanding issues: setting good default starting values for smooth terms better printing of output (hide fixed terms? print smooth terms in a way more consistent with mg...
github.com
October 12, 2026 at 9:04 PM
tedpavlic.bsky.social
New widget for exploring GLMM's or warm-starting #stats coding after quick prototyping:
tpavlic.github.io/topic_visual...

Import your own data or use sample data sets. Try different response families/random effects. Export to #Rstats code (+MATLAB/Python). Translate classic tests to GLMM.
GLMM Analyzer – generalized linear mixed models in the browser
Fit generalized linear models, and mixed models with one grouping factor, to a table in the browser: normal, binomial, Poisson, negative binomial, gamma, beta, and ordinal responses with interactions ...
tpavlic.github.io
October 12, 2026 at 12:39 AM
paperposterbot.bsky.social
arXiv📈🤖
Trustworthy Method Comparison with AI Judges: Estimation and Design under Order, Batch, and Aggregation Effects
By Li, Ding
Large language models (LLMs) are increasingly used as judges for automated AI evaluation. A common practice is to randomize prompt sequences and average the resulting scores, but its statistical validity remains unclear. We show that LLM evaluation mechanisms can be approximated by a class of Markov generalized linear mixed models (GLMMs), supported by out-of-sample predictions across three major commercial LLMs. Using a first-order Markov GLMM, we study leaderboard ranking and group comparison. For leaderboard ranking, randomize-and-average selection is consistent under a mild separation condition, and a Williams square design can improve efficiency when item qualities are close. For group comparison, naive averaging can yield inconsistent conclusions about differences in group-level quality because of the response model's nonlinearity. Empirical results further support the validity of the proposed model-based inference beyond the first-order theory, including settings with higher-order sequence memory. We illustrate the approach in an application where AI judges compare two graphical model estimation methods.
October 7, 2026 at 11:49 AM
mattansb.msbstats.info
This is also why "knowing to code" means different things to different people - different levels of abstraction are required for different tasks.

For fitting a GLMM I don't need to know how memory is allocated or even how different optimizers find the MLE..
October 1, 2026 at 4:24 AM
arxivecongnbot.bsky.social
arXiv📈🤖
Copula based joint regression models for correlated data: an analysis in the bivariate case
By
arXiv:2410.11892v1 Announce Type: new Abstract: Regression analysis of non-normal correlated data is commonly performed using generalized linear mixed models (GLMM) and generalized estimating equations (GEE). The recent development of generalized joint regression models (GJRM) presents an alternative to these approaches by using copulas to flexibly model response variables and their dependence structures. This paper provides a simulation study that compares the GJRM with alternative methods. We focus on the case of the marginal distributions having the same form, for example, in models for longitudinal data. We find that for the normal model with identity link, all models provide accurate estimates of the parameters of interest. However, for non-normal models and when a non-identity link function is used, GLMMs in general provide biased estimates of marginal model parameters with inaccurately low standard errors. GLMM bias is more pronounced when the marginal distributions are more skewed or highly correlated. However, in the case that a GLMM parameter is estimated independently of the random effect term, we show it is possible to extract accurate parameter estimates, shown for a longitudinal time parameter with a logarithmic link model. In contrast, we find that GJRM and GEE provide unbiased estimates for all parameters with accurate standard errors when using a logarithmic link. In addition, we show that GJRM provides a model fit comparable to GLMM. In a real-world study of doctor visits, we further demonstrate that the GJRM provides better model fits than a comparable GEE or GLM, due to its greater flexibility in choice of marginal distribution and copula fit to dependence structures. We conclude that the GJRM provides a superior approach to current popular models for analysis of non-normal correlated data.
October 1, 2026 at 8:49 PM
josetermita351.bsky.social
youtube.com/watch?v=glMm...
Rajoy: "Cuanto peor para todos, mejor. Mejor para mí el suyo beneficio político"
YouTube video by elDiarioes
youtube.com
September 22, 2026 at 1:03 PM
plantdiseasej.bsky.social
Findings from Pawan Basnet et al. highlight that continuous planting of soybean containing resistance genes can lead to shifts in soybean cyst #nematode populations toward more virulent types. Learn more: https://doi.org/10.1094/PDIS-09-25-1939-RE
Fig. 3. GLMM effects for SCN female index across different HG type indicator lines and experimental years. * Significant at P ≤ 0.05 ** Significant at P ≤ 0.01 *** Significant at P ≤ 0.001.
September 14, 2026 at 9:20 AM
paperposterbot.bsky.social
arXiv📈🤖
A Fast Monte Carlo Newton-Raphson Algorithm to Estimate Generalized Linear Mixed Models with Dense Covariance
By Watson, Wang, Giorgi
Estimation of Generalised linear mixed models (GLMM) including spatial Gaussian process models is often considered computationally impractical for even moderately sized datasets. In this article, we propose a fast Monte Carlo maximum likelihood (MCML) algorithm for the estimation of GLMMs. The algorithm is a stochastic Newton-Raphson method, which approximates the expected Hessian and gradient of the log-likelihood by drawing samples of the random effects. We propose a new stopping criterion for efficient termination and preventing long runs of sampling in the stationary post-convergence phase of the algorithm and discuss Monte Carlo sample size choice. We run a series of simulation comparisons of spatial statistical models alongside the popular integrated nested Laplacian approximation method and demonstrate potential for similar or improved estimator performance and reduced running times. We also consider scaling of the algorithms to large datasets and demonstrate a greater than 100-fold reduction in running times using modern GPU hardware to illustrate the feasibility of full maximum likelihood methods with big spatial datasets.
September 13, 2026 at 6:35 PM
cranupdates.bsky.social
New on CRAN: soReta (0.1.0). View at https://CRAN.R-project.org/package=soReta
soReta: Ready-to-Analyze Datasets from Camera Trap Data
Functions to build datasets ready for statistical analysis from camera trap data: GLMM/GAMM on counts/RAI at various temporal levels, group size, occupancy, kernel/circular analysis of activity patterns, temporal interactions between species, hierarchical diel models, and classic capture-mark-recapture. Input data must be formatted in the style produced by 'camtrapR', the standard convention in the field.
CRAN.R-project.org
September 12, 2026 at 7:00 PM
cranberriesfeed.bsky.social
New CRAN package soReta with initial version 0.1.0
#rstats
https://cran.r-project.org/package=soReta
CRAN: Package soReta
Functions to build datasets ready for statistical analysis from camera trap data: GLMM/GAMM on counts/RAI at various temporal levels, group size, occupancy, kernel/circular analysis of activity patterns, temporal interactions between species, hierarchical diel models, and classic capture-mark-recapture. Input data must be formatted in the style produced by 'camtrapR', the standard convention in the field.
cran.r-project.org
September 12, 2026 at 3:03 PM
gerrythill.bsky.social
StrainSpy’s GLMM framework enables straightforward integration of large, multi-cohort microbiome studies. Re-analysis of a large colorectal cancer dataset identified both known and novel strain-level associations, including in E. coli, with an accompanying enrichment of adhesin-encoding genes.
September 2, 2026 at 12:56 PM
josetermita351.bsky.social
youtube.com/watch?v=glMm...
Rajoy: "Cuanto peor para todos, mejor. Mejor para mí el suyo beneficio político"
YouTube video by elDiarioes
youtube.com
August 30, 2026 at 6:38 PM
josetermita351.bsky.social
youtube.com/watch?v=glMm...
Rajoy: "Cuanto peor para todos, mejor. Mejor para mí el suyo beneficio político"
YouTube video by elDiarioes
youtube.com
August 29, 2026 at 9:24 PM
mzloteanu.bsky.social
#statstab #604 Prediction can be safely used as a proxy for explanation in causally consistent Bayesian generalized linear models

Thoughts: Interesting take on causal inference & model misspecification.

#brms #causalinference #bayesian #glmm #prediction

www.tandfonline.com/doi/full/10....
www.tandfonline.com
August 27, 2026 at 3:57 PM
tapani-hopkins.bsky.social
Vielä kun tajuaisi miten nuo glm glmm jne tuon tekevät. Laskeminen minulta vielä luonnistuisi (aloitin teoreettisen fysiikan opiskelijana), mutta joskus tuntuu että tilastolliset ohjelmat ja menetelmät ovat edenneet niin nopeasti, että useimmat vain tökkäävät näppäintä ja toivovat tulosten pitävän.😅
August 23, 2026 at 5:18 PM
tomospotter.bsky.social
It seems like you might want a GLMM, with random intercepts for trap. Something like glmmTMB(n_wasps ~ forest_type + (1|trap_id)+ (1|species), family=“poisson”) perhaps? This would account for non-independence among samples from a given trap, and non-independence among samples for each species.
August 23, 2026 at 4:03 PM
tiinasavikangas.bsky.social
Meneekö se ns random effectinä (vrt muut prediktorit fixed effectejä), eli olisi nimenomaan glmm? Sitä olisin tarjonnut pitkittäisaineistoon, mutta kai se toimii myös yhden aikapisteen vasteelle 😔 Kyllä ChatGPT auttaa koodin kanssa 😁 Mutta voin huomenna vilkaista omia skriptejä kans.
August 23, 2026 at 6:14 PM
aaronstevenwhite.io
The solution to one's blood sugar being in the boolean algebra is indeed generally Tanqueray 10 and a walk. post-like.exe.xyz
Spent the last hour trying to map a GLMM into a new algebra homomorphism, only to realize I forgot to eat lunch. The code compiles, but my blood sugar is in the boolean algebra now. Need a Tanqueray 10 and maybe a walk around Rochester before I touch the shot logger again.
August 19, 2026 at 3:47 PM