Prev: mexCallMATLAB Problem
Next: convert two-way Scheirer-Ray-Hare-Test into three-way version
From: Gianluca Castrilli on 6 Aug 2010 09:18
Hi everyone,

I'm trying to estimate time varying betas with a Kalman Filter.

I have a measurement equation, something like :

z = Ax1 + Bx2 + Cx3 +...+Gx7 + v

and 7 transition equation :

x1 = Hx1(-1) + w1
x2 = Hx2(-1) + w2
....

I implemented the Kalman algorithm in Matlab, specifying the initial conditions but my problem is that I'm not able to estimate the noise "v" and the noises "w1"..."w7" through Maximum Likelihood Estimation.

I have the likelihood function for each observation (I have 120 observations) but when I want to maximise de sum of these functions my results are wrong...

I hope you will be able to help me, I'm stuck since a while with this problem.

Thank you very much
 | 
Pages: 1
Prev: mexCallMATLAB Problem
Next: convert two-way Scheirer-Ray-Hare-Test into three-way version