geneseq: Generation aof random sequences
In sequence: Analysis of Sequences of Events

Description Usage Arguments Details Value Note Author(s) See Also Examples

Generates nseq random sequences of items represented by codes, either by random sampling (order 0), either using a Discrete Time Markov Chain (DTMC) transition matrix (order 1).

1	geneseq(nseq, lmin, lmax, order = 0, dico = NULL, mattrans = NULL)

`nseq`	numeric: number of sequences to generate.
`lmin`	numeric: minimum length of sequences
`lmax`	numeric: maximum length of sequences
`order`	numeric: order of the Markov Process, 0 or 1. See details.
`dico`	character vector: dictionnary of items.
`mattrans`	data.frame: transition matrix with identifiers as first column.

Generates random sequences of states according to a Markov process of order 0 or 1.
order 0: the following state is drawn with equal probabilities in the item list, no influence of previous state.
order 1: the following state is drawn with a probability proportional to the product of the transition matrix by the vector of states.
nseq sequences of random size are generated. The size of each sequence is a random integer drawn under uniform probability between lmin and lmax (both included).

A list of nseq items

`A1`	list: first sequence
`A2`	list: second sequence

... and so on until nseq.

The transition matrix must be square and Markovian. A non-square matrix produces an error and program break. Its columns must all sum to 1. No verification is done for this last point. A matrix in which the columns do not sum to 1 may lead to spurious or unpredictible results.

Jean-Sebastien Pierre
Jean-sebastien.pierre@univ-rennes1.fr

compseq

# 1 - generates 5 random sequences of length from 1 to 15 with  
# symbols a,b,c,d.(order = 0)
geneseq(5,1,15,dico=letters[1:4]) 
mat<-data.frame(id=letters[1:3],a=c(0,0.75,0.25),b=c(0.5,0.25,0.25),
c=c(0.1,0.8,0.1))
# 2 - generates 2 random sequences of length 10 with symbols a,b,c,d  
# according to the transition matrix mat
geneseq(2,10,10,order=1,mattrans=mat)