English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Zendy Plus

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Zendy Tools

Zendy Open

A method is introduced for approximate marginal likelihood inference viaadaptive Gaussian quadrature in mixed models with a single grouping factor. Thecore technical contribution is an algorithm for computing the exact gradient ofthe approximate log-marginal likelihood. This leads to efficient maximumlikelihood via quasi-Newton optimization that is demonstrated to be faster thanexisting approaches based on finite-differenced gradients or derivative-freeoptimization. The method is specialized to Bernoulli mixed models withmultivariate, correlated Gaussian random effects; here computations areperformed using an inverse log-Cholesky parameterization of the Gaussiandensity that involves no matrix decomposition during model fitting, while Waldconfidence intervals are provided for variance parameters on the originalscale. Simulations give evidence of these intervals attaining nominal coverageif enough quadrature points are used, for data comprised of a large number ofvery small groups exhibiting large between-group heterogeneity. The Laplaceapproximation is well-known to give especially poor coverage and high bias fordata comprised of a large number of small groups. Adaptive quadrature mitigatesthis, and the methods in this paper improve the computational feasibility ofthis more accurate method. All results may be reproduced using code availableat \url{https://github.com/awstringer1/aghmm-paper-code}.

Exact Gradient Evaluation for Adaptive Quadrature Approximate Marginal Likelihood in Mixed Models for Grouped Data