[R] stochastic process transition probabilities estimation

Massimo Santini santini at dsi.unimi.it
Tue Apr 4 13:41:41 CEST 2000

Hi all,
	I'm new with R (and S), and relatively new to statistics (I'm a
computer scientist), so I ask sorry in advance if my question is silly. 

My problem is this: I have a (sample of a) discrete time stochastic
process {X_t} and I want to estimate 

 Pr{ X_t | X_{t-l_1}, X_{t-l_2}, ..., X_{t-l_k} } 

where l_1, l_2, ..., l_k are some fixed time lags. It will be enough for
me to compute 

 #{ X_t=a_t, X_{t-l_1}=a_{t-l_1}, X_{t-l_2}=a_{t-l_2}, ..., X_{t-l_k}=a_{t-l_k} }
          #{ X_{t-l_1}=a_{t-l_1}, X_{t-l_2}=a_{t-l_2}, ..., X_{t-l_k}=a_{t-l_k} }

for any given sequence of a_t, a_{t-l_1}, a_{t-l_2}, ..., a_{t-l_k}.

If I should do this in C, Fortran or some other algol-like programming language
I've no doubt it will take to me no more than few minutes to write down the code. 

But I'm interested in doing it within R framework in order to further
use such a result in a statistical context, where other powerfull
tools are available. Shurly I can try to implement an external
function and use it with R, but this probably wouldn't be as simple as
to write the code in R itself.

An idea should be to use R in a "multivariate" fashion: if I can get
factors of a multivariate random variable, I can build a random vector
Y_t using X_t, and its lags: Y_t = [ X_t, X_{t-l_1}, X_{t-l_2}, ...,
X_{t-l_k} ], and taking the factors of Y_t and of its margin without
the first component I think I'm done.  Unfortunately, factor, sort,
unique all work only with univariate data.

Any other suggestion will be very appreciated...

Best regards, Massimo

Dr.  Massimo Santini       
Tel. Milano +390255006278, Crema +390373898227;  Fax. +390255006373       
<mailto:santini at dsi.unimi.it> <http://zorn.usr.dsi.unimi/~santini/>

r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch

More information about the R-help mailing list