Trouble is, they require that I provide a posterior log-likelihood function, which is a simple enough requirement that seems to be giving me problems for some reason.<BR><BR>I know it should be ...

In this example, the log likelihood function of the SSM is computed using prediction error decomposition. The annual real GNP series, y t, can be decomposed as where ...

and is the normal probability function. This is the likelihood function for a binary probit model. This likelihood is strictly positive so that you can take a square root of and use this as your ...

through optimization of the following log-likelihood function: We show that this approach permits the use of initial reference maps that are devoid of any prior information (or bias) about the ...