dtermining the significance in heterogeniety in effect sizes when conducting a three-level meta-analytics models


Pleae help in this issue.. I am conducting three-level meta-analytic models. when I am trying to determine the significance of heterogeneity in effect sizes it gives me this message:

  •                v, 
  •                random = list(~ 1 | id2, 
  •                              ~ 1 | id1), 
  •                sigma2=c (0,NA),
  •                tdist = TRUE, 
  •                data = HSV1)

Error in .ll.rma.mv(opt.res$par, reml = reml, Y = Y, M = V, A = A, X.fit = X, :
Final variance-covariance matrix not positive definite.
In addition: Warning messages:
1: In rma.mv(y, v, random = list(~1 | id2, ~1 | id1), sigma2 = c(0, :
There are outcomes with non-positive sampling variances.
2: In rma.mv(y, v, random = list(~1 | id2, ~1 | id1), sigma2 = c(0, : 'V' appears to be not positive definite.
How do i fix this issue?
Thank you for any help get

Hi, and welcome!

Please see the FAQ: What's a reproducible example (`reprex`) and how do I do one? Using a reprex, complete with representative data will attract quicker and more answers.

Could you edit your post to allow others to see how this comes about? It's much easier to offer answers without having first to reverse engineer the issue.

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.