DEL19D Model

# Cut-andPaste Code Below into Window Above and Run

#

# DEL19D BAU Model

#

# Measurement Matrix (Growth-P) (LS-P) (P+LS-GD)

#

#            Q      LS        W       P        N       PD        G      GD

#[1,]  0.40810 0.01687  0.40549 -0.1733  0.40434  0.40637  0.38881  0.3986

#[2,] -0.08042 0.74396 -0.02061 -0.6438 -0.10690 -0.07661 -0.07337  0.0500

#[3,]  0.07865 0.65971  0.12278  0.7173  0.01306  0.11209  0.06332 -0.1107

# Fraction of Variance 

#[1] 0.7380 0.9254 0.9737 0.9893 0.9954 0.9985 0.9998 1.0000

#

require(dse)

require(matlab)

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

f <- matrix( c(  1.00469302, -0.064160984, 0.01610147, 0.24021238,

                      -0.09378164,  0.784883642, 0.08343431, 0.07238104,

                        0.01946481, -0.007641104, 0.84078485, 0.02460858,

                   0.000000000,  0.00000000,  0.0000000, 1.00000000

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

#

# To Stabilize the System, Uncomment Next line

# f[1,1] <- 0.90

#

h <- eye(3,4)

k <- f[1:4,1:3,drop=FALSE]

DEL19 <- SS(F=f,H=h,K=k,z0=c(0.24021238, 0.07238104, 0.02460858, 1.00000000),

              output.names=c("DE1","DE2","DE3"))

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

#tfplot(simulate(DEL19,sampleT=20))

shockDecomposition(toSSChol(DEL19))

DEL19.data <- simulate(DEL19,sampleT=50,start=1872,noise=matrix(0,50,3))

m <- l(DEL19,DEL19.data)

tfplot(m)

AIC(m)

tfplot(forecast(m,horizon=50))

#

# Feedback Components

#

DEL19f <- SS(F=f[2:4,2:4,drop=FALSE],H=eye(2,3),K=f[2:4,2:3,drop=FALSE],

z0=c( 0.03947626, 0.01232109, 1.00000000),

              output.names=c("DE2","DE3"))

stability(m <- SS(F=f[2:3,2:3,drop=FALSE],H=eye(2),Q=eye(2),R=eye(2)))

shockDecomposition(m)