[R] EM algorithm

Elena 5/12 elena_verbruggen at hotmail.com
Wed Jul 22 00:31:37 CEST 2009


l=∑_i∑_k log [(Se(i))^(t_ik ) [1-Se(i) ]^(1-t_ik ) p+(Sp(i))^(1-t_ik )
[1-Sp(i) ]^(t_ik ) (1-p)]

This is my loglikelihood. It is a sum over i (I tests) and over k (K
patients). t_ik is the test results for the ithe test for the kth patient. 

Se(i)=sensitivity of test i, Sp(i)=specificity of test i and p=prevalence
are my three unknown parameters. Now I need to apply the EM algorithm to
this l to become estimates for these parameters. But I am not able to code
it in R, it does not work.

Do you know how to code this likelihood and the EM algorithm?
-- 
View this message in context: http://www.nabble.com/EM-algorithm-tp24596990p24596990.html
Sent from the R help mailing list archive at Nabble.com.




More information about the R-help mailing list