Error in t(y) %*% x : non-conformable arguments for Summary function

Hi! I am getting "Error in t(y) %*% x : non-conformable arguments" when I use the summary function for my GMM model.

GMM <- pgmm(log_revenue ~ lag(log_revenue, 1:1) + CSat + Engage + Innov
            |lag(log_revenue, 2:2) | lag(CSat, 1:1) + lag(Engage, 1:1) + lag(Innov,1:1),
            data =, effect = "twoways",model = "twosteps",
            collapse = TRUE,transformation = "ld", fsm = "full",
            subset = sample == 1, time.dummies = "TRUE")

How can I solve this issue?
I did use this code before and it was working back then. But I am experiencing this problem now.
Also, it works well using random/fixed effect estimations.

Do all the example documented in ?pgmm work for you ?
If no, that would be interesting and might reveal something.

If yes, then I would think we need to see your or something similar, or else how should we reproduce your issue ?

FAQ: How to do a minimal reproducible example ( reprex ) for beginners

Yes, they all work.
I do get the result of the summary function for fixed/random effect models on this data.
I am also getting the result of the summary function for GMM using another dataset.
I get the same error when I want to run the following code:

mtest(GMM, order = 2, vcov = NULL)

It does also work for order = 1, but not for order = 2

so about that reprex?

I figured the problem is related to AR(2). I only have 4 years of data, whereas in order to have the AR(2) 5 years is needed. Since "summary" couldn't report AR(2), the error was coming up.

Thank you for following up on the issue.

This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.