[R] problem with eigen() function

[Patrick Burns]

This problem reminds me of the problem described in:

tolstoy.newcastle.edu.au/*R*/devel/03b/1304.html

In that case a specific matrix causes 'eigen' to return
NaNs in the eigen vectors.  But dumping the matrix
and reading it back in does not reproduce the problem.
That you are getting a hang rather than merely garbage
returned is considerably more annoying I would imagine.


Patrick Burns
patrick at burns-stat.com
+44 (0)20 8525 0696
http://www.burns-stat.com
(home of S Poetry and "A Guide for the Unwilling S User")

Jingqin luo wrote:

>Dear R-users,
>   Recently, I have come across a weird problem. I run a large number of iterations and at one of the step within each iteration, I calculate the eigen values of a updated covariance matrix. From all my intermediate output, the code freezes after printing out the covariance matrix but before printing out the eigen values. So, obviously it stops at the only step, the eigen() function calculation. However, it didn't give any error message from this eigen function. Also, when I copy the printed out covariance matrix over and calculate its eigen function, there is no problem of giving the answer. So, nothing is wrong with this covariance matrix either. I suspect that probably during these iteration, I have returned too many objects or R had sth wrong with the memory allocation or ...... I have no idea how to debug it so far since I can't go into the eigen function step by step each time is run. If someone has any similar experience or good advice on debugging this problem, I
> would really appreciate it. Thanks!
>  Jingqin
>
> 
>---------------------------------
>It's here! Your new message!
>Get new email alerts with the free Yahoo! Toolbar.
>	[[alternative HTML version deleted]]
>
>______________________________________________
>R-help at stat.math.ethz.ch mailing list
>https://stat.ethz.ch/mailman/listinfo/r-help
>PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
>and provide commented, minimal, self-contained, reproducible code.
>
>
>  
>