[SciPy-user] Help required regding Eigenvectors

[Fernando Perez]

Sudheer Phani wrote:
> Hello All,
> 
>    I have been encountering some strange problems with the calculation of
> eigen vectors using python. The problem is with the sign of the eigen
> vectors calculated. On Very few occasions the eigen vectors calculated
> using python have an opposite sign than the actual eigenvectors, though
> the magnitude is correct.

Eigenvectors are defined only up to a scalar normalization constant.  What do 
you mean exactly by opposite sign?  If x is an eigenvector for

A*x = s*x  (s being the eigenvalue)

then obviously

A*(-x) = s*(-x)

so -x is also an eigenvector with the same eigenvalue (or a.x for a being any 
scalar).

Best,

f