Doctoral thesis

Statistical Inference on Network Data: Spatial Panel and Latent Variables

ContributorsJiang, Chaonan
Defense date2021-06-03

This thesis develops new models and inference methods for network structures, and contains two parts. In the first part, we develop new higher-order asymptotic techniques for the Gaussian maximum likelihood estimator in a spatial panel data model, with fixed effects, time-varying covariates, and spatially correlated errors. Our saddlepoint density approximation features relative error of order O(1/(n(T-1))) with n being the cross-sectional dimension and T the time-series dimension. Monte Carlo experiments illustrate the good performance of our approximation over first-order asymptotics and Edgeworth expansions. In the second part, we derive Laplace-approximated maximum likelihood estimators (GLAMLEs) of parameters in our Graph Generalized Linear Latent Variable Models. Then, we study the statistical properties of GLAMLEs and display the estimation results in a Monte Carlo simulation considering different numbers of latent variables. Moreover, we make a comparison between Laplace and variational approximations for inference of our model.

  • Network data analysis
  • Higher-order asymptotics
  • Investment-saving
  • Random field
  • Tail area
  • Latent variable models
  • Laplace approximations
Citation (ISO format)
JIANG, Chaonan. Statistical Inference on Network Data: Spatial Panel and Latent Variables. 2021. doi: 10.13097/archive-ouverte/unige:152558
Main files (1)

Technical informations

Creation06/17/2021 4:44:00 PM
First validation06/17/2021 4:44:00 PM
Update time03/07/2024 8:28:06 AM
Status update03/07/2024 8:28:06 AM
Last indexation03/07/2024 8:28:08 AM
All rights reserved by Archive ouverte UNIGE and the University of GenevaunigeBlack