Hi all
I write below code for simulation and forecasting a stochastic mortality
model with parameters kappa2(t) and gamma3(t-x) where t is the time and x
the age but I got this message:"Error in G3.sim[, k] = G3.aug :
number of items to replace is not a multiple of replacement length"
Any suggestion ?
a=arima(gamma3,c(1,1,0))
D2=diff(kappa2)
mu2=mean(D2)
nD2=length(D2)
sigma2=sum((D2-mu2)^2)/nD2
sigma2=sqrt(sigma2)
Dg=diff(gamma3)
mug=mean(Dg)
nDg=length(Dg)
sigmag=sum((Dg-mug)^2)/nDg
sigmag=sqrt(sigmag)
t0=1
t1=length(y)
t2=t1+tmax
y.aug=(t0:t2)+y[1]-1
lx=length(x)
c=(y[1]-x[lx]):(y[t1]-x[1])
ct1=length(c)
ct2=ct1+tmax
c.aug=(y.aug[1]-x[lx]):(y.aug[t1]-x[lx])
K2.sim=array(0,c(length(y.aug),N))
G3.sim=array(0,c(length(c.aug),N))
mtx.sim.array=array(0,c(length(y.aug),lx,N))
#mtx=years x ages x simulations
for(k in 1:N){
K2.aug=(t0:t2)*0
K2.aug[t0:t1]=kappa2
K2.aug[(t1+1):t2]=K2.aug[t1]+cumsum(rnorm(tmax,mu2,sigma2))
K2.sim[,k]=K2.aug
G3.aug=0
G3.aug[t0:ct1]=gamma3
G3.aug[(ct1+1):ct2]=G3.aug[ct1]+mug
+(a$coef)*(G3.aug[ct1]-G3.aug[ct1-1]-mug)+cumsum(rnorm(tmax,0,sigmag))
G3.sim[,k]=G3.aug
for(i in 1:lx){
mtx.sim.array[,i,k]=exp(beta1[i]+beta2[i]*K2.aug+G3.aug)
}
}
list(x=x,y=y,y.aug=y.aug,dtx=dtx,etx=etx,wtx=wtx,beta1=beta1,beta2=beta2,
kappa2=K2.sim,gamma3=G3.sim,mtx=mtx.sim.array)
}
Thanks for your time and your help
--
View this message in context:
http://r.789695.n4.nabble.com/forecasting-number-of-items-to-replace-is-not-a-multiple-of-replacement-length-error-tp3705100p3705100.html
Sent from the R help mailing list archive at Nabble.com.
______________________________________________
[email protected] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
