The MARSS package in R offers function for dynamic factor analysis. In this package, the dynamic factor model is written as a special form of state space model and they assume the common trends follow AR(1) process. As I am not very familiar with those two methods, I come with two questions:

Is the Dynamic Factor Analysis a special form of State Space Model? What is the difference between those two methods?

In addition, the Dynamic Factor Analysis does not necessary assume the common trends as AR(1) process. Is there any package that allows the the common trends as seasonal ARIMA (or some other) process?


1 Answer 1


I did not see your question before.

Yes, dynamic factor analysis can bee seen as a particular case of state-space model. It makes observations dependent of a small dimensional state vector (small relative to the dimension of the observation vector). So it is the same idea as in ordinary factor analysis, plus time dependence.

The "factors" may have any time dynamics. Several R packages, if you use R, will let you specify a general dynamic factor analysis model, including for instance dlm or KFAS.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.