[R] Calculating transition probabilties

David Winsemius dwinsemius at comcast.net
Tue May 13 23:02:04 CEST 2014


On May 13, 2014, at 7:49 AM, Baba Bukar wrote:

> Dear all,
> 
> I am new to R and have some problem computing transition probabilities. My
> problem goes like this;
> 
> data_set <-
> c(2,0,45,6,78,3,0,2,6,0,5,8,0,2,8,9,12,212,22,4,1,0,3,5,88,5,69,12,4,0,0,0,0,4,87,6,99,104,22,7)
> 
> observations greater than, say 3, is considered as useful (denoted as 1)
> while less than 3 are not useful (denoted as 1). Am trying to calculate the
> transition in these count data such as P_1,1=prob from useful to useful,
> P_1,0=prob from useful to not useful, P_0,1=prob from not useful to useful
> and P_0,0=prob from not useful to not useful.
> 

The rle function (run length encoding) might be helpful here:

> rle( data_set >= 3 )$lengths
 [1] 2 4 2 1 1 2 2 6 2 7 4 7
> rle( data_set >= 3 )$values
 [1] FALSE  TRUE FALSE  TRUE FALSE  TRUE FALSE  TRUE FALSE  TRUE FALSE  TRUE

> Thank you much as you respond soonest
> 
> Kind regards
> Zakir
> 
> 	[[alternative HTML version deleted]]

This is a plain text mailing list.

-- 
David Winsemius
Alameda, CA, USA



More information about the R-help mailing list