A COMPARISON OF PSEUDO-MAXIMUM LIKELIHOOD AND ASYMPTOTICALLY DISTRIBUTION-FREE DYNAMIC FACTOR-ANALYSIS PARAMETER-ESTIMATION IN FITTING COVARIANCE-STRUCTURE MODELS TO BLOCK-TOEPLITZ MATRICES REPRESENTING SINGLE-SUBJECT MULTIVARIATE TIME-SERIES

The study of intraindividual variability pervades empirical inquiry in virtually all subdisciplines of psychology. The statistical analysis of multivariate time-series data - a central product of intraindividua l investigations - requires special modeling techniques. The dynamic f actor model (DFM), which is a generalization of the traditional common factor model, has been proposed by Molenaar (1985) for systematically extracting information from multivariate time-series via latent varia ble modeling. Implementation of the DFM model has taken several forms, one of which involves specifying it as a covariance-structure model a nd estimating its parameters from a block-Toeplitz matrix derived from the multivariate time-series. We compare two methods for estimating D FM parameters within a covariance-structure framework - pseudo-Maximum Likelihood (p-ML) and Asymptotically Distribution Free (ADF) estimati on - by means of a Monte Carlo simulation. Both methods appear to give consistent model parameter estimates of comparable precision, but onl y the ADF method gives standard errors and chi-square statistics that appear to be consistent. The relative ordering of the values of all es timates appears to be very similar across methods. When the manifest t ime-series is relatively short, the two methods appear to perform abou t equally well.