You are here

Missing values error without missings in the data?

3 posts / 0 new
Last post
nastjuscha's picture
Offline
Joined: 06/28/2016 - 04:51
Missing values error without missings in the data?
AttachmentSize
Binary Data DT_FEM.R712 bytes
File dt.dat8.27 KB

Dear Mike,

I'm afraid I have to ask you for help again.
I receive a following new error warning every time I try to run an FEM regardless of the constellation of variables from my dataset:
> summary(fixed1)
Error in if (pchisq(chi.squared, df = df, ncp = 0) >= upper) { :
missing value where TRUE/FALSE needed
Even if I try to run it only with first three variables (there are 146 full matrices without any missings) I have the same problem.
I appreciate any idea you can provide.

Regards,
Nastja

Mike Cheung's picture
Offline
Joined: 10/08/2009 - 22:37
Dear Nastja, The issue is

Dear Nastja,

The issue is that the lengths of dt and nDT are different (146 vs. 147). tssem1() fails to detect it in this case. I have fixed it in the developmental version in github. It will throw an error. When I drop the last element in nDT, it works fine.

By the way, the fixed-effects model takes a long time to run as there are 146 groups. Moreover, there is a large degree of heterogeneity on the correlation matrices. The random-effects model seems to be better. It only takes a few seconds to run.

Best,
Mike

nastjuscha's picture
Offline
Joined: 06/28/2016 - 04:51
Hi Mike, such a silly

Hi Mike,

such a silly mistake! Thank you very much!

Regards,
Nastja