Grocer Function
NAME
tvp_param1ad - transformation of tvp parameters into kalman compatible ones
CALLING SEQUENCE
[Q,R]=tvp_param1ad(param)
PARAMETERS
Input
-
param = vector of parameters
-
k = size of the vectors of parameters
Output
-
Q = variance of the state equation
-
R = variance of the observation equation
-
b0 = (k x 1) vector of intial conditions for b
DESCRIPTION
In a time-varying estimation, transforms the vectors parameters into their matrix and vectors counterpart in the corresponding Kalman filter.
NOTE: Q is not assumed diagonal and priorb0 is not estimated. Parameters are the original ones (not squared as in tvp_param1 or tvp_param1a).
EXAMPLE
This function should have no other use than providing the transformation of the parameters that must be estimated into the matrices used in the kalman filter.
AUTHOR
Eric Dubois 2002