Neural composite likelihood estimation: simulation based inference for time series
Abstract
Simulation based inference (SBI) circumvents the challenge of intractable likelihoods by using a simulator that generates data given parameter values. For instance, neural likelihood estimation (NLE) estimates the likelihood function by training a neural network to perform conditional density estimation on simulated data given corresponding parameters. However such density estimation is only feasible for relatively low dimensional data. We extend the scalability of SBI methods to a higher dimensional problem: long sequences with a complex dependency structure. We introduce Neural Composite Likelihood Estimation (NCLE). This divides the sequence into smaller, equal-sized batches. Instead of training NLE to estimate the likelihood for an entire sequence, we estimate the likelihood for each batch separately. The product of these forms an approximate composite likelihood (CL), and we perform frequentist inference using methods from the CL literature: we get a point estimate from maximising the approximate CL and obtain confidence intervals by estimating the Godambe information matrix. We demonstrate the effectiveness of NCLE with experiments on time series models.
Related
- Neural Estimation of Summary Statistics for Simulation-Based Inference with Non-Asymptotic Guarantees
- Simulation based-inference of epidemiological and phylodynamic models via Neural Posterior Estimation
- Marginal Composite Likelihood Estimation of Bergomi Models from High-Frequency Option Prices
- Simulation-consistent Estimation of the Marginal Likelihood for Block Models
- A perspective note on likelihood approximation and inference for complex simulation models using a chain of aggregated normalizing flows
BibTeX
@misc{yan2026neural,
title = {{Neural composite likelihood estimation: simulation based inference for time series}},
author = {Grace Yan and Mark Beaumont and Dennis Prangle},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.21762},
eprint = {2609.21762},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.21762},
}
RIS
TY - UNPB
TI - Neural composite likelihood estimation: simulation based inference for time series
AU - Grace Yan
AU - Mark Beaumont
AU - Dennis Prangle
PY - 2026
DA - 2026/09/18
PB - arXiv
DO - 10.48550/arxiv.2609.21762
UR - https://arxiv.org/abs/2609.21762
AB - Simulation based inference (SBI) circumvents the challenge of intractable likelihoods by using a simulator that generates data given parameter values. For instance, neural likelihood estimation (NLE) estimates the likelihood function by training a neural network to perform conditional density estimation on simulated data given corresponding parameters. However such density estimation is only feasible for relatively low dimensional data. We extend the scalability of SBI methods to a higher dimensional problem: long sequences with a complex dependency structure. We introduce Neural Composite Likelihood Estimation (NCLE). This divides the sequence into smaller, equal-sized batches. Instead of training NLE to estimate the likelihood for an entire sequence, we estimate the likelihood for each batch separately. The product of these forms an approximate composite likelihood (CL), and we perform frequentist inference using methods from the CL literature: we get a point estimate from maximising the approximate CL and obtain confidence intervals by estimating the Godambe information matrix. We demonstrate the effectiveness of NCLE with experiments on time series models.
ER -