A0319
Title: MSAEM estimation for multidimensional four-parameter normal ogive models
Authors: Xiangbin Meng - Northeast Normal University (China) [presenting]
Abstract: A mixed stochastic approximation expectation maximization (MSAEM) algorithm coupled with a Gibbs sampler is developed to compute the marginalized maximum a posteriori estimate (MMAPE) of a multidimensional four-parameter normal ogive (M4PNO) model. The proposed MSAEM algorithm has the computational advantages of the stochastic approximation expectation maximization (SAEM) algorithm for multidimensional data. It also alleviates the potential instability caused by label switching and improves the estimation accuracy. Simulation studies are conducted to illustrate the good performance of the proposed MSAEM method, where MSAEM consistently performs better than SAEM and some other existing methods in multidimensional item response theory. Moreover, the proposed method is applied to a real data set from the 2018 Programme for International Student Assessment (PISA) to demonstrate the usefulness of the 4PNO model and MSAEM in practice.