Smith Controller

AIC <- function(model) {informationTestsCalculations(model)[3]}

require(dse)

require(matlab)

#

# Measurement Matrix 

#         K      L      Q

#[1,] 0.577  0.577  0.577

#[2,] 0.812 -0.334 -0.478

#

# Fraction of Variance 

#[1] 1 1 1

#

f <- matrix( c(-6.331224e-17,    0,  2.705009e-01,

                       1.405814e-31,    0 , 1.912512e-16,

                       0.000000e+00,    0,  1.000000e+00

),byrow=TRUE,nrow=3,ncol=3)

k <- (f[1:3,1:2,drop=FALSE])

h <-matrix( c(1,    0,    0,

                      0,    1,    0

),byrow=TRUE,nrow=2,ncol=3)

Smith <- SS(F=f,H=h,K=k,z0=c(2.705009e-01, 1.912512e-16,1),

              output.names=c("Smith1","Smith2"))

print(Smith)

is.SS(Smith)

stability(Smith)

# tfplot(simulate(Smith,sampleT=100))

Smith.data <- simulate(Smith,sampleT=100,start=1)

Smith.f <- forecast(l(Smith,Smith.data),horizon=100)

tfplot(Smith.f)

Smith_RW <- SS(F=eye(2),H=eye(2),K=eye(2),z0=c(2.705009e-01, 1.912512e-16),

              output.names=c("Smith1","Smith2"))

tfplot(simulate(Smith_RW,sampleT=100,start=1))

shockDecomposition(Smith_RW)

AIC(l(Smith_RW,Smith.data))