WE_E20 Model

# Cut-and-Paste Code Below into Window Above an Run

#

# WE_E20 Model

#

# Measurement Matrix (overall-HOURS-XREAL), (XREAL-X), (Q-N-L-U)

#

#         Q      N      U   HOURS   XREAL       X      L

#[1,] 0.231  0.418  0.422 -0.4280 -0.3203  0.4024  0.382

#[2,] 0.297  0.248  0.180  0.0687  0.6937 -0.3924  0.422

#[3,] 0.909 -0.186 -0.217 -0.0863 -0.0981 -0.0114 -0.274

#

 #Fraction of Variance 

#[1] 0.748 0.883 0.993 1.000 1.000 1.000 1.000

#

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

require(dse)

require(matlab)

f <- matrix( c(1.00629959, -0.05766949, -0.09542032, 0.12654495,

                      0.04296649,  1.06699880,  0.27025187, 0.10806276,

                       0.05938140, -0.01188805,  1.04875201, 0.02649331,

0.00000000,  0.0000000,  0.0000000,  1.0000000000

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

#

# To create RWf model, uncomment following line

# f[1,1] <- f[2,2] <- f[3,3] <- 1.0

#

h <- eye(3,4)

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

WE_E20 <- SS(F=f,H=h,K=k,z0=c( 0.12654495, 0.10806276, 0.02649331,  1.0000000000),

              output.names=c("WE1","WE2","WE3"))

print(WE_E20)

is.SS(WE_E20)

stability(WE_E20)

 WE_E20.data <- simulate(WE_E20,sampleT=150,start=1900)

# WE_E20.data <- simulate(WE_E20,sampleT=20,noise=matrix(0,20,3))

WE_E20.f <- forecast(l(WE_E20,WE_E20.data),horizon=50)

tfplot(WE_E20.f)

AIC(l(WE_E20,WE_E20.data))

shockDecomposition(toSSChol(WE_E20),horizon=10,shock=rep(-1,10))