Initial State

questions concerning analysis/theory using programs M-SURGE, E-SURGE and U-CARE

Initial State

Postby Luke Wakeen » Fri Jul 27, 2012 4:45 pm


I am wondering if it is necessary to do a model selection for the Initial State in models that don't have uncertainty in the initial state. I am told that even though a model's initial condition may be given, the AIC is lower when the initial state is not held constant. Does anyone know why this might be and what the mathematical justification for it would be?

Luke Wakeen
Posts: 3
Joined: Fri Jul 13, 2012 3:48 pm

Re: Initial State

Postby CHOQUET » Mon Jul 30, 2012 9:04 am

When there is no uncertainty in the IS then the likelihood can be decomposed in two distinct parts.
As a consequence, MLE of the other parameters are independent, i.e. you don't need to model IS.
See section 2.8. of the manual of E-SURGE
Posts: 211
Joined: Thu Nov 24, 2005 4:58 am
Location: CEFE, Montpellier, FRANCE.

Return to analysis help

Who is online

Users browsing this forum: No registered users and 1 guest