The full paper is available by request.
Four methods for estimating a dynamic factor model, the direct autoregressive factor
score (DAFS) model, are evaluated and compared. The ¯rst method estimates the DAFS
model using a Kalman filter algorithm based on its state space model representation. The
second one employs the maximum likelihood estimation method based on the construction
of a block-Toeplitz covariance matrix in SEM the framework. The third method is built in
the Bayesian framework and implemented using Gibbs sampling. The fourth one is the
least square method which also employs the block-Toeplitz matrix. All four methods are
implemented in currently available software. The simulation study shows that all four
methods reach appropriate parameter estimates with comparable precision. Differences
between the four estimation methods and related software are discussed.
A copy of the manuscripts can be obtained by email.
Suggestion or comment can be sent to firstname.lastname@example.org.