A0179
Title: Bayesian analysis of nonlinear structured latent factor models using a Gaussian process prior
Authors: Yimang Zhang - SUSTECH (China)
Jian Qing Shi - SUSTECH (China) [presenting]
Xiaorui Wang - SUSTECH (China)
Abstract: Factor analysis models are widely utilized in social and behavioral sciences, such as psychology, education, and marketing, to measure unobservable latent traits. A nonlinear structured factor analysis (FA) model is introduced, which is more flexible in characterizing the relationship between manifest variables and latent factors, and then the confirmatory identifiability of the latent factor is given to ensure the substantive interpretation of the latent factors. A Bayesian approach with a Gaussian process prior is proposed to estimate the unknown nonlinear function. Asymptotic results are established, including structural identifiability of the latent factors, consistency of all parameters and the unknown nonlinear function. Simulation studies and real data analysis are conducted to investigate the performance of the proposed method. Simulation studies and real data analysis show the proposed method performs well in handling nonlinear model and successfully identifies the latent factors.
