Fast inference in generalized linear models via expected log-likelihoods
From MaRDI portal
Abstract: Generalized linear models play an essential role in a wide variety of statistical applications. This paper discusses an approximation of the likelihood in these models that can greatly facilitate computation. The basic idea is to replace a sum that appears in the exact log-likelihood by an expectation over the model covariates; the resulting "expected log-likelihood" can in many cases be computed significantly faster than the exact log-likelihood. In many neuroscience experiments the distribution over model covariates is controlled by the experimenter and the expected log-likelihood approximation becomes particularly useful; for example, estimators based on maximizing this expected log-likelihood (or a penalized version thereof) can often be obtained with orders of magnitude computational savings compared to the exact maximum likelihood estimators. A risk analysis establishes that these maximum EL estimators often come with little cost in accuracy (and in some cases even improved accuracy) compared to standard maximum likelihood estimates. Finally, we find that these methods can significantly decrease the computation time of marginal likelihood calculations for model selection and of Markov chain Monte Carlo methods for sampling from the posterior parameter distribution. We illustrate our results by applying these methods to a computationally-challenging dataset of neural spike trains obtained via large-scale multi-electrode recordings in the primate retina.
Recommendations
- Fast maximum likelihood estimation using continuous-time neural point process models
- Maximum Likelihood Algorithms for Generalized Linear Mixed Models
- scientific article; zbMATH DE number 774851
- Expectation propagation for likelihood-free inference
- A sequential reduction method for inference in generalized linear mixed models
Cites work
- scientific article; zbMATH DE number 3945130 (Why is no real title available?)
- scientific article; zbMATH DE number 45785 (Why is no real title available?)
- scientific article; zbMATH DE number 3626409 (Why is no real title available?)
- scientific article; zbMATH DE number 1220667 (Why is no real title available?)
- scientific article; zbMATH DE number 2107836 (Why is no real title available?)
- scientific article; zbMATH DE number 2117879 (Why is no real title available?)
- scientific article; zbMATH DE number 961607 (Why is no real title available?)
- A new look at state-space models for neural data
- Adaptive Forward-Backward Greedy Algorithm for Learning Sparse Representations
- Applied multivariate statistical analysis.
- Asymptotic Statistics
- Asymptotics of graphical projection pursuit
- Bayes Factors
- DISTRIBUTION OF EIGENVALUES FOR SOME SETS OF RANDOM MATRICES
- Efficient methods for sampling spike trains in networks of coupled neurons
- Generalized Cross-Validation as a Method for Choosing a Good Ridge Parameter
- Hierarchical models for assessing variability among functions
- Introductory lectures on convex optimization. A basic course.
- MCMC using Hamiltonian dynamics
- Maximum likelihood analysis of spike trains of interacting nerve cells
- Modeling the impact of common noise inputs on the network activity of retinal ganglion cells
- Pattern recognition and machine learning.
- Sequential Optimal Design of Neurophysiology Experiments
- Spline smoothing: The equivalent variable kernel method
- Understanding Gaussian Process Regression Using the Equivalent Kernel
Cited in
(8)- Fast Generalized Linear Models by Database Sampling and One-Step Polishing
- Autoregressive Point Processes as Latent State-Space Models: A Moment-Closure Approach to Fluctuations and Autocorrelations
- Learning neural connectivity from firing activity: efficient algorithms with provable guarantees on topology
- Inference of a mesoscopic population model from population spike trains
- Faster model matrix crossproducts for large generalized linear models with discretized covariates
- Reconstruction of recurrent synaptic connectivity of thousands of neurons from simulated spiking activity
- Fast maximum likelihood estimation using continuous-time neural point process models
- Approximate maximum likelihood estimation in cure models using aggregated data, with application to HPV vaccine completion
This page was built for publication: Fast inference in generalized linear models via expected log-likelihoods
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1704759)
