#

# W_E20 Model Cut-and-Paste Code below into window above and Run

#

# Measurement Matrix  (Growth-XREAL), (XREAL+Q-T), (T+XREAL-L)

#          Q      N  XREAL       X       L      T

#[1,] 0.3934  0.471 -0.232  0.4614  0.4743  0.363

#[2,] 0.5140  0.146  0.742  0.0112  0.0842 -0.395

#[3,] 0.0375 -0.192  0.477 -0.0230 -0.2188  0.828

#

# Fraction of Variance 

#[1] 0.717 0.925 0.982 1.000 1.000 1.000

#

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

require(dse)

require(matlab)

f <- matrix( c( 1.002900989, -0.01221580, -0.3389237,  0.11196869,

                       0.089905713,  1.06700279,  0.3800768,  0.07620913,

                      -0.001940628,  0.07357085,  0.2758457, -0.02539065,

0.00000000,  0.0000000,  0.0000000,  1.0000000000

),byrow=TRUE,nrow=4,ncol=4)

#

# To Stabilize Model, Uncomment Next line

# f[1,1] <- f[2,2] <- 0.9

#

h <- eye(3,4)

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

WE20 <- SS(F=f,H=h,K=k,z0=c(0.11196869,  0.07620913, -0.02539065,  1.0000000000),

              output.names=c("W1","W2","W3"))

print(WE20)

is.SS(WE20)

stability(SS(F=f[1:3,1:3,drop=FALSE],H=eye(3),R=eye(3),Q=eye(3)))

# tfplot(simulate(WE20,sampleT=100))

WE20.data <- simulate(WE20,sampleT=100,noise=matrix(0,100,3),start=1900)

WE20.f <- forecast(l(WE20,WE20.data),horizon=100)

tfplot(WE20.f)

AIC(m <- l(WE20,WE20.data))

shockDecomposition(m)