[Limdep Nlogit List] Error components model with nonsigificant asc's but significant standard deviations

[Stephen Beville]

Greetings

With an RPL-error components model how should an analyst handle
model estimates where the standard deviations of latent random effects are
statistically significant but the non-random asc estimates are not
statistically significant?

In other words if the asc's are nonsignificant-does this imply that error
component estimates of the standard deviations of latent random effects are
imaterial?

Thank you,

Stephen Beville
PhD Candidate
Environment Society & Design Division
Lincoln University
(027) 338 8438
<*{:):):):><{