What is wrong when do the chapman-Klmogorov test

Construction and validation of Markov models.
POST REPLY
User avatar
aurum bai
Posts: 1
Joined: Mon Dec 21, 2009 5:00 am

What is wrong when do the chapman-Klmogorov test

Post by aurum bai » Mon Apr 22, 2013 7:04 am

Hi admin,

When I do the Markov using EMMA, I meet one wrong report. I do know the main reason of this err. Maybe the model is not good.

Could you give me some suggestion?

The err as below:

Code: Select all

./mm_example_chapman 
Command line parameters valid.
Sparse matrix with dimension ( 50 x 50 ) read successfully from file './matrix/transitionmatrixOf_Traj1_Traj2.ascii'.
Calculating stationary distribution.
Exception in thread "main" java.lang.RuntimeException: Real component 18 of stationary distribution is not positive.
	at sherlock.stochasticprocesses.markov.impl.StationaryDistributionFromTransitionMatrix.perform(StationaryDistributionFromTransitionMatrix.java:109)
	at sherlock.emma.cmd.ChapmanKolmogorovTestCmd.run(ChapmanKolmogorovTestCmd.java:184)
	at de.martin_senne.jcommandline.AbstractCmd.check(AbstractCmd.java:33)
	at de.martin_senne.jcommandline.AbstractCmd.go(AbstractCmd.java:45)
	at sherlock.emma.cmd.ChapmanKolmogorovTestCmd.main(ChapmanKolmogorovTestCmd.java:339)
Top
POST REPLY