Announcement

Collapse
No announcement yet.
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • LRTest in MLM with Sample Weights

    Hello Everyone

    I want to compare two MLM models:

    Code:
     
     gen ID=_n  reshape long DVW, i(ID) j(group)
    Code:
    Model 1: 
     mixed DVW c.CV1 i.CV2 i.group c.IV1 c.IV2 c.IV [pw=weight3] || ID:  c.IV1 c.IV2 c.IV, estimates store A
    Code:
    Model 2: 
      mixed DVW c.CV1 i.CV2 i.group c.IV1##i.group c.IV2##i.group c.IV##i.group [pw=weight3]|| ID:  c.IV1 c.IV2 c.IV, estimates store B
    Code:
    lrtest A B
    However, I get the error message: "LR test likely invalid for models with robust VCE"

    Is this because of the sample weights? If so, how do I compare these two models with the sample weights included? Or, should I compare models without sample weights, and then, after identifying the most parsimonious model, estimate with sample weights?

    Any help is appreciated.

    Joe


Working...
X