DE_LM Model

#

#

# DE_LM Model (Germany)

#

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

require(dse)

require(matlab)

#

# Measurement Matrix (Growth-HOURS), (L+N-X), (X+L-U)   

#                                   (EXPORT LED GROWTH?)

#          Q       N        U  HOURS   XREAL      X     L

#[1,]  0.389  0.3728  0.39069 -0.374  0.3867  0.368 0.364

#[2,]  0.157  0.4491  0.00261  0.434 -0.2155 -0.493 0.544

#[3,] -0.170 -0.0985 -0.48120  0.244 -0.0912  0.649 0.491

#

# Fraction of Variance 

#[1] 0.934 0.999 1.000 1.000 1.000 1.000 1.000

#

f <- matrix( c( 

                      0.9995837694, 0.03504777, -0.001153864,  0.168405987,

                     -0.0323463858, 1.00764178, -0.380964610, -0.006413983,

                     -0.0003978753, 0.02765853,  0.991209573, -0.008411533,

                      0.000000000,  0.00000000,  0.0000000, 1.000000000

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

#

# To Stabilize System Uncomment Next Line

# f[1,1] <- 0.9895879317; f[2,2] <- 0.99756536; f[3,3] <- 0.981297478

#

h <- eye(3,4)

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

DE_LM <- SS(F=f,H=h,K=k,z0=c(0.33652679, -0.01511981, -0.01699353,  1.0000000000),

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

print(DE_LM)

is.SS(DE_LM)

stability(DE_LM)

# tfplot(simulate(DE_LM,sampleT=100,start=1950))

DE_LM.data <- simulate(DE_LM,sampleT=100,noise=matrix(0,100,3),start=1950)

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

tfplot(DE_LM.f)

AIC(m)

shockDecomposition(toSSChol(DE_LM))