Rose,
1- Inside your loop, you have to use F1[i], not F[i] (2 occurences). By the
way, never use F (or T) as variables in R, because it also means FALSE and TRUE.
2- Your P update must be at the end of the loop. You haven't calculated your
K[i] yet where you placed it.
y<-(Nile) n.y<-length(y) P<-numeric(n.y) a<-numeric(n.y) i<-y
P[1]<-10000000 a[1]<-0 n<-100 sigma.2.epsilon<-15099
sigma.2.eta<-1469.1 nv <- matrix(nrow=100, ncol=1)
F1<-P[1]+sigma.2.epsilon K <-P[1]/F1 for (i in 2:n-1){
F1[i]<-P[i]+sigma.2.epsilon K[i]<-P[i]/F1[i] nv[i]<-y[i]-a[i] a[i+1]
<-a[i]+K[i]*nv[i] P[i+1]<-P[i]*(1-K[i])+sigma.2.eta }
Neuro> Date: Wed, 21 Feb 2007 15:41:48 -0800> From: rbentosa@yahoo.com>
To: r-help@stat.math.ethz.ch> Subject: [R] loops in R help me please> >
I am trying to make the following Kalman filter equations work and therefore
produce their graphs.> v_t=y_t - a_t> a_t+1=a_t+K_t*v_t>
F_t=P_t+sigma.squared.epsilon> P_t+1=P_t*(1-K_t)+sigma.squared.eta>
K_t=P_t/F_t> > Given:> a_1=0,P_1=10^7,sigma.squared.epsilon=15099,>
sigma.squared.eta=1469.1> > I have attached my code,which of course doesnt
work.It produces NAs for the Fs,Ks and the a.> Can somebody tell me please
what am I doing wrong in this loop?> Why doesnt this loop work as it should
be;to produce plots of a declining filtered state variance(P_t),prediction
errors(v_t),and a declining prediction variance(F_t)?> Also,how can I
construct and plot the 90% confidence intervals around a_t?> Thanks for your
help and look forward.> Leonard> > > >
---------------------------------> It's here! Your new message!> Get
new email alerts with the free Yahoo! Toolbar.
_________________________________________________________________
Soyez parmi les premiers à essayer Windows Live Mail.
4911fb2b2e6d
[[alternative HTML version deleted]]