A simulation study on computation and inference accuracy of factor loadings for large data mines

Abstract

This study simulates and examines the weighting effects of using relatively diminutive samples to infer a gigantic data mine. The statistical inference in the simulations is carried out by using a basic factor analysis model. Several stratum sizes and stratum-vs.-population ratios are designed in the data generating procedures; therefore, the experiment can reflect the practical research environments. Estimations are conducted by using the altered maximum likelihood estimation (MLE) algorithm. The preliminary results show the proposed method is promising.