#

# AXJ_LM Model (cut-and-paste code into window above and Run (Cmd-Enter)

#

#

# Measurement Matrix (Growth), (N-XREAL), (Q-XREAL)

#           Q       N   XREAL

#[1,]  0.5809  0.5736  0.5775

#[2,] -0.2243  0.7949 -0.5638

#[3,]  0.7825 -0.1979 -0.5904

#

# Fraction of Variance 

#[1] 0.9857 1.0000 1.0000

#

merge.forecast <- function (fx,n=1) {

#

# Merges a forecast with the outputdata

x <- splice(fx$pred,fx$forecast[[n]])

colnames(x) <- seriesNames(fx$data$output)

return(x)

}

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

require(dse)

require(matlab)

f <- matrix( c(1.032483086, 0.0893329688,  7.3532892,  1.146422e-01,

                     -0.015620695, 0.9745223725, -6.5969415,  1.761040e-03,

                      0.000010184, 0.0007955634,  0.8679133, -7.008574e-05,

0.00000000,  0.0000000,  0.0000000,  1.0000000000

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

#

# Stabilize Asian Growth

# (Uncommment next command)

#

#       f[1,1] <- 1

#

h <- eye(3,4)

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

AXJLM <- SS(F=f,H=h,K=k,z0=c( 1.146422e-01,  1.761040e-03, -7.008574e-05,  1.0000000000),

              output.names=c("AXJ1","AXJ1","AXJ1"))

print(AXJLM)

is.SS(AXJLM)

stability(AXJLM)

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

AXJLM.data <- simulate(AXJLM,sampleT=100,noise=matrix(0,100,3))

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

AXJM.fx <- merge.forecast(AXJLM.f)

tfplot(AXJLM.f)

AIC(l(AXJLM,AXJLM.data))

q <- matrix( c( 0.3075393229, 4.394617e-04,  3.442278e-05,

                        0.0576136724, 2.775657e-03, -3.088211e-05,

                       -0.0002702703, 2.256986e-05,  4.062942e-06,

0.00000000,  0.0000000,  0.0000000

),byrow=TRUE,nrow=4,ncol=3)

r <- matrix( c( 0.2951994354, 0.000000e+00, 0.000000e+00,

                       0.0613396050, 3.005610e-03, 0.000000e+00,

                      -0.0003710925, 2.324968e-05, 4.681277e-06

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

AXJLM2 <- SS(F=f,H=h,Q=q,R=r,z0=c( 1.146422e-01,  1.761040e-03, -7.008574e-05,  1.0000000000),

              output.names=c("AXJ1","AXJ1","AXJ1"))

shockDecomposition(AXJLM2)

#

# WL20 India

#

#

# India IN_LM Model

#

require(dse)

require(matlab)

#

# Measurement Matrix (Growth) (N-Q)

#           Q      N

#[1,]  0.7071 0.7071

#[2,] -0.7071 0.7071

#

# Fraction of Variance 

#[1] 0.9919 1.0000

#

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

f <- matrix( c(  2.416284e-13,  4.538441e-13,  2.227908e-14,

                      -4.026590e-14, -9.208321e-14, -3.864347e-15,

                      0.0000000, 0.0000000,  1.0000000

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

g <- matrix(c(0.81867254, 0.2053109, 3.308165,

                    -0.00223055, 0.6147680, 1.405612,

                     0.00000000, 0.0000000, 0.000000

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

h <- eye(2,3)

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

INLM <- SS(F=f,H=h,K=k,z0=c( -1.6951995, -0.2577112,  1.0000000),

              output.names=c("Growth","N-Q"))

stability(INLM)

#tfplot(simulate(INLM,sampleT=100,noise=matrix(0,100,2),start=1,input=AXJM.fx))

INLM.data <- simulate(INLM,sampleT=100,start=1,input=AXJM.fx)

m <- l(INLM,INLM.data)

#tfplot(m)

INLM.f <- forecast(m,horizon=100,input=AXJM.fx)

tfplot(INLM.f)

AIC(m)